Skip to main content

Full text of "Prehistoric And Protohistoric Mysore"

See other formats







Assistant Professor of History and Archaeology 

in the 

University of Mysore 





Hie study of the stone-using cultures of pre- and proto-historic Mysore 
herein presented is intended initially to survey the evidence that has so far 
become available for the interior of Southern India. While the boundaries 
of Mysore State have been accepted as limiting the study from the 
geographical point of view, it has been necessary to transgress these political 
boundaries from time to time in order to include material which has an 
obvious bearing on the past of Mysore. 

It must be admitted that the results of the present analysis of the material 
are modest and that they show up the gaps in our present knowledge rather 
than dosing them. This is inevitable, however, in the case of an investigation 
of this kind and has the advantage of showing where special work should be 
done in the future. 

From the chronological point of view this study covers the earliest period 
of mankind in Mysore, namely the Palaeolithic Period, and those periods 
during which microlithic stone tools were used. It is well known that at 
least pan of the microlithic period is contemporary with the use of metal, 
but so little attention has been paid to the stone tool assemblages of the 
cultures in question that a considerable portion of the time involved in writing 
this book was spent on their study. Valuable publications have already 
appeared setting forth the evidence connected with other archaeological 
materials, the most important being Sir Mortimer Wheeler’s on Brahmagiri 
and Chandravalli, and for this reason such evidence has been relegated to 
the background. 

Throughout this present work the relation of stone tool-making man 
to his natural environment has been continually kept in mind. Stress has 
been laid on two obvious factors in this respect, namely the geographical 
situation of the sites and the availability of natural raw materials. 

In addition to the collections of the Department of Environmental 
Archaeology, London University, which contain artifacts from numerous 
Indian sites, including a large number of microlithic and “ neolithic ” 
specimens presented by Col. D. H. Gordon, a study has been made of the 
Mysore material housed in the Government Museum, Madras, the 
Kibbanahalli material kept in the Geology Museum of the Central College, 
Bangalore, the British Museum, the Victoria and ^Albert Museum and a 
collection of specimens from Brahmagiri made by the writer. 

In the course of the preparation of this work several frieodfc have helped 
me. First and foremost, I owe a great debt of gratitude to Professor F. E. 
Zeuner who initiated me into the subject of geochronology in 1949, while he 
was in India by the special invitation of the Government of India to carry 
out a survey of prehistoric sites. Further training under him in London, 
especially in Stone Age typology, was possible owing to the generosity of the 
British Council to whom I express my sincere gratitude for their travel 
grant. My thanks are also due to Professor K. de B. Codrington, head of 
the Indian Department of the Institute of Archaeology, London University. 

I am greatly indebted to the following people who have been kind 
enough to allow me to examine the specimens included in this work: 
Professors L. Rama Rao and B. S. Madhava Rao (Central College, Bangalore); 
Messrs. E. M. M. Alexander (British Museum) and J. Irwin (Victoria and 

Albert Museum); and Mrs. B. Allchin (London). In addition Dr. A. 
Aiyappan (Madras Museum) kindly provided photographs of specimens from 
die Foote collection. I gratefully acknowledge the help I have received at 
their hands. 

A number of people have sacrificed much valuable time in preparing 
the manuscript for the press and offering criticism and help, including Dr. 
I. W. Cornwall, Miss J. T. Philips and Miss J. Sheldon. 

Finally, I wish to express my gratitude to the Central Research Fund 
Committee of London University for a grant in aid of research work and to 
Professor V. L. D’Souza and other members of the Publication Committee 
of the Mysore University for a grant which has made the publication of this 
volume possible. In substance it is a greatly modified version of a much 
larger thesis submitted for the Doctorate in Philosophy of London University 
in 1952. Those requiring detailed information, especially typological tables 
analysing the material from various sites, should consult the thesis. 



Li$t of contents Page 

Chapter I. Environmental Background 

A. General considerations 1 

B. The setting of die scene 1 

C. Exploitation of die environment 7 

Chapter II. The Palaeolithic industries of Mysore 

A. The Palaeolithic of India 12 

B. Palaeolithic sites in Mysore 14 

C. An analysis of the Mysore Palaeolithic industries IS 

D. Typological comparison of the Mysore Palaeolithic 

industries with those from other parts of India 24 

Chapter III. Microlithic industries of Mysore 

A. Introduction 27 

B. Microlithic sites of India 28 

C. Microlithic sites of Mysore 29 

O. Classification of microlithic industries from Mysore 31 

E. Differences between the Jalahalli and Brahmagiri 

industries 36 

F. Differences between the Jalahalli and Kibbanahalli 

industries 37 

G. Comparable sites outside Mysore 37 

H. Chronology 40 

Chapter IV. The “neolithic” complex in Mysore 

A. Introduction 44 

B. Chalcolithic or Neolithic ? SO 

C. Polished stone implements 51 

D. Technique of making celts 52 

E. Axe types from Brahmagiri 54 

F. Modes of hafting Brahmagiri axes and adzes 55 

G. Chronology 57 

Chapter V. The Megalithdc problem in Mysore 

A. Megaliths in relation to raw material 59 

B. Geographical location of sites 59 

G Megaliduc architecture 64 

D. Grave goods 66 

E. The age of the megaliths 70 

Bibliography 73 

List of Hates 82 

Plates 86 


Chapter I 



The study of the natural environment is of great importance to the 
prehistorian. If he wishes to understand a culture as a function of mankind, 
the environment to which it was an adaptation must be reconstructed. In 
fact, it is one of the cardinal principles of human geography to trace the 
dose connection that exists between physical regions and cultural areas. 
The distribution of culture is literally based on rock (Richards 1929). 

During the Palaeolithic Period environment ruled man with an iron 
hand. Not only the extent of the dependence of culture on environment 
but also the doseness of correlation between them were greatest in the lower 
stages of cultural growth. In the subsequent stages man succeeded in mak- 
ing himsdf increasingly independent of local environment and at the same 
time exploiting it more efficiently. 

Prehistoric archaeology may be regarded as ethnography projected back- 
ward in time. Environmental problems call for an intimate contact with 
the natural sciences, on which prehistoric archaeology relies entirely, both 
for chronological purposes and for establishing the environmental conditions 
that obtained during the particular period under consideration. 

From the chronological point of view, however, the geological evidence 
of Mysore has not yet been made available. Work like that recently done 
by Zeuner (1950) in Gujarat is still wanting in southern India. But it is 
already possible to take advantage of geological and geographical observa- 
tions, combining them with the distribution and material evidence of the 
stone-using cultures, in order to obtain some information about the de- 
pendence of the prehistoric people of Mysore on their environment. 


Physiography. The State of Mysore is a well defined physical entity 

in the south of India. It is a table-land situated in the angle where the 

eastern and western ghats respectively converge towards the Nilgiri-hills 
and has been not inaptly described as a rocky triangle. The area extends 
between the parallels of 11° 36' and 15°2' north latitude and between the 
meridians of 74°4Q' and 78° 36' east longitude embracing an area of 

29,474.82 square miles (excluding the Bellary region). This political limit 
is to some extent an accident ol history and therefore should not strictly 
confine us to the above-mentioned area. Both in the prehistoric and historic 
periods a continuity of cultures prevailed in the contiguous and neighbouring 
territories, particularly of southern Bellary, a fact which is confirmed by 

The general elevation rises from about 2,000 ft. above sea level along 
the northern and southern frontiers to about 3,000 ft. along the central 
water-parting, which separates the basin of the Kistna (Krishna) from that 


of the Cauvery (Kaveri) 1 and divides the country into nearly equal 
parts. These are further longitudinally intersected by hills, single or in 
chains, running parallel to the coasts. These hills are caused by the geo- 
logical structure of the country, which is characterised by N-S trends, and 
they appear at one time to have been linked with the Annamalai, Palni, 
Travancore and Shevaroy hills, die eventual separation from them being 
due to marine denudation (Oldham, 1893), The landscape is undoubtedly 
of great antiquity, the Pleistocene having contributed only minor features. 
The broad open valleys, the low gradients of the rivers and the gentle 
undulating aspect of the country are characteristic of such an ancient land 

Mysore falls into two distinct physical regions called Malnad and 
Maidan. The Malnad or hill country lies to the west, bordering the western 
ghats and comprises the Shimoga, Chikmaglur and parts of the Hassan 
District. It is a region where the beautiful and the picturesque meet; of mag- 
nificent hills and forests. Rainfall is heavy and the vegetation belongs to 
the evergreen type. 

As opposed to the Malnad, the Maidan or open country with its plains 
of alluvial black soil and rice-fields comprises a vast area including the 
Mysore, Bangalore, Tumkur, Kolar, Mandya and Chitaldrug districts. Here 
the rainfall is moderate with correspondingly open vegetation. It is not 
surprising that the Maidan bears the densest human occupation in Mysore 
at the present day, for thick evergreen forests with heavy rainfall do not 
lend themselves to the development of high civilization. The same condition 
appears to have obtained in the past. No prehistoric sites are known from 
the Malnad, except for a few Palaeolithic ones in the Maidan borderland. 
Palaeolithic man, being particularly attracted to river valleys, appears to 
have penetrated into the Malnad to some extent. It is, however, by no 
means certain that the climate was the same in Palaeolithic times. The 
evidence concerning this problem at present available from Mysore is 
too scanty. 

Precipitation. Mysore is an area of extremes as regards rainfall. Although 
over the whole state the average rainfall does not exceed 36 inches, Agumbi 
in the Shimoga District receives 317 inches and Nayakanahatti, in the Chital- 
drug District only 16 inches — the highest and the lowest rainfall areas 
respectively. In 4 out of the 9 districts, it ranges from 25 inches to 30 
inches. The rainfall figures clearly indicate that 

(a) the climate is wet in Malnad, and 

(b) dry in the Maidan or open country, Chitaldrug District being 
the driest. 

A consideration of rainfall and climate should interest students of 
human cultures because they are responsible for movement and migration of 
peoples and cultures (Brooks, 1949, p.9). There is a movement of people 
from regions which are naturally moist to regions which are naturally dry 
during a period of increased rainfall. In drier periods the direction of the 
movement is reversed. Peake (1922, p. 151) found it necessary to call in 
the aid of climatic fluctuations in order to understand the migration* of the 

(!) The old spelling of these rivers has been kept, although incorrect, to avoid confusion. 


Aryans, whom he called ‘ wires,’ Huntington (1924) believed that the rise 
and decline of the Maya Civilization of Yucatan could only be explained by 
climatic changes. 

It seems possible that neolithic man of Brahmagiri, in the Chitaldrug 
District, had a damper climate, for in the lotfer layers of the excavations at 
that place there occurred a series of water-logged deposits, containing large 
quantities of snail shells. 

The major river systems. The study of the river systems is of the utmost 
importance to the prehistorian, since prehistoric chronology is closely linked 
up with river chronology. The Thames in England, the Vezere and Somme 
in France, the Hm-Saale system of central Germany are some of the classic 
European examples. Rivers have, at all times, provided man with water 
supply and lines of communication. In addition, they supplied him with 
his food. He could watch the animals that came to drink, and hunt them. 
Furthermore, the pebbles of the rivers constituted an important raw material 
for the manufacture of his tools. It is in these river terraces that we find 
large quantities of the discarded implements of Palaeolithic man. He lived 
in the valleys which provided him with raw materials for implements and 
kept a constant watch on the river banks for his prey. The well-known 
concentration of Palaeolithic sites in the river valleys thus proves. to be real, 
and is not a fortuitous result of collection. As plate III shows, the majority 
of the Palaeolithic sites in Mysore are indeed situated in the valleys of the 
large rivers or their tributaries. 

Later, with the development of agriculture in the Neolithic period, the 
river valleys became centres of man’s activity in a different sense. For 
agricultural activities, proximity to the water supply was important, but 
equally essential was the availability of more or less level, arable surfaces. 
Hence, the Neolithic' 1 > and later agricultural settlements are often situated 
on river flood-plains. No wonder that the early civilizations in Egypt, India, 
China and Mesopotamia arose in river valleys, where the floodplains were 
wide and easily cultivated, and where seasonal flooding brought moisture 
and a suspension of fresh soil. The flooding restricts the settlements to 
artificial hills or to the sides of the valleys. In Mysore, Brahmagiri is 
a case in point, though on a small scale, since it is situated at the foot of 
a hill near a stream which even today floods the floodplain occasionally. 
It is natural, therefore, that river valleys play a prominent part in Indian 
prehistory. The Indus and the Soan in die Punjab, the Sabarmati in Gujarat, 
the Narbada in central India, the Godavari, the Kortallaiyar, the Palar and 
the Tungabhadra in southern India are outstanding examples. 

There are three main river systems in Mysore, namely the Cauvery in the 
south, the Tungabhadra in the north and the two Pennars and the Palar 
in the east. 

The Mysore portion of the course of the Ttingabhadra is a very im- 
portant river of the central Malnad. It is composed of the twin rivers Tunga 
and Bhadra, both of which rise at a place called Gangamula in the western 
Ghats. After being fed by a number of streams like the Haridra, the 

(I) In referring to the Neolithic as applied to the Near East or Europe a capital “ N ” is used, but in India 
a small lerter is used both for neolithic and microltthic since they arc not regarded as periods. 


Vedavati (Haggari), and the Kumudvati, it eventually joints the Kistna. 
Beyond the Mysore frontier, the Tungabhadra becomes a Maidan river. 
Bruce Foote (1916, p. 10) explored its banks beyond Mysore, and discovered 
at Hampasagara a veiy fine trap celt and megalithic and Andhra pottery. 
More recently, Palaeolithic tools were found at the same place by Dr. Subba 

This site was visited by the writer in 1950, when he was a member of 
Professor Zeuner’s expedition. Since no Palaeolithic sites have yet been 
stratigraphically studied in Mysore State itself and since Hampasagar lies 
only a short distance from the state boundary, it may be described here as 
an example of such site. The exact location is east of Gauri Nalla, where a 
double terrace is seen. The section extending to the higher level shows 
loose gravel on rock, covered by kankarised gravel and about 10 feet of 
kankarised silt. This carries a soil section with a grey A-horizon (c.2 feet) 
and a brown B-horizon (c.7 feet). 

The palaeoliths came from the loose gravel, which is only a few feet 
thick. The thinness of these Palaeolithic gravels is a feature common to 
most rivers in the interior of South India. It means that they owe their 
existence not to an actual aggradation, but to the lining of the bed of the 
river of that time. Similar conditions were observed by Zeuner (1950) in 
Gujarat. The modern rivers have a similar pebble lining which is moved 
when the river is in flood but which lies dry and accessible to man in the 
dry season. This explains the large number of palaeoliths found is such 
situations: pebbles were collected and converted into implements on the 
spot. Many were of course rolled by the river later on. The implication 
is that the climatic conditions in the Palaeolithic need not have been very 
different from those of today. The climate was certainly not wet enough 
for laterite to be formed, and a dry season occurred regularly. 

Foote states that he found “later Neolithic or early Iron Age cemeteries” 
in what is here described as the B-horizon. Both this material and palaeoliths 
are bound to be incorporated in the modern river, where the implements 
are frequently found in a rolled condition. 

Mention must also be made of the Chinna Hagari basin, in the north- 
east comer of Mysore, a Maidan area, belonging to the Tungabhadra system 
The Chinna Hagari, a small river, is a tributary of the Hagari river, which 
flows through northern Mysore into the Bellary District. In its valley 
lie many historic and prehistoric sites, of which Siddapur, Jettingerameswar 
and Brahmagiri lie in Mysore, whilst many others, including Gudekallu, 
Hosahalli, Rayadrug, Mallapuram, Gallapalli, and Addaguppa belong to the 
Bellary District. But they all form a single cultural unit and belong to the 
microlithic complex. 

The Cauvery (Kaveri) in the south of Mysore forms an entity by 
itself. Rising in Coorg, it flows through South Mysore till it abruptly quits 
the tableland at Sivasamudram. Some interesting historic pottery comes 
from T. Narasipur on its bank. At Lakshampura, 7 miles from T. 
Narasipur, on the left bank of the same river, pottery of the early historic 
period was found by Foote. Three or four miles from the Cauvery, similar 
pottery came to light at a place called French-rocks. Since both series 
contain sherds with graffiti, they appear to be of megalithic origin. 


The Pennar, Ponnaiyar and the Palar are the rivers of south-eastern 
Mysore. The Palar plain beyond the Mysore boundary was examined Dy 
Foote (1873) in the early sixties of the last century. Further study of the 
Palar basin was carried out, with a view to elucidating climatic changes in 
south-east India in the Palaeolithic, by Richards (1932). 

Geological environment. From this point of view, India is divided into three 
geological divisions, namely : 

(1) the Peninsular shield 

(2) the Indo-Gangetic plain 

(3) the Himalayan Belt. 

We are concerned only with the first part here (Plate I). 

In the classification of the rocks of Peninsular India the word ‘ Archaean’ 
is commonly used. The term was introduced by J. G. Dana in 1872 to 
designate formations older than the Cambrian. In America its use is now 
restricted to the highly metamorphosed schistose, gneissic and granite rocks, 
while the term ‘Algonkian' is used to include undoubted original sediments 
lying below the base of the Cambrian. In India, the formations below the 
eparchaean (epi-archaean) unconformity have been included by Sir Thomas 
Holland under Archaean. Within the Archaean, a lower transition system 
(identified with the so-called Dharwars of Foote) and an upper transition sy- 
stem are distinguised but the Archeans and the Dharwar system are not differ- 
ent units, since the granites and the gneisses may merely represent certain 
horizons within the schistose members. This at any rate is the view of 
Dr. Krishnan (1949, p. 96). 

Foote was the first to study the Dharwars in the eighteen-eighties. Dr. 
Smeeth’s view that the Dharwars of Mysore are of igneous origin or the 
view of B. Rama Rao (1940) of the Mysore Geological Survey that some 
of them are undoubtedly of sedimentary origin need not concern us in this 
context. It suffices to know that the Dharwars have provided numerous 
raw materials for early man. 

The Dharwars occur in Mysore in three geographic groups. They 
constitute a corrugated mantle, with a strike of approximately N-S, which 
conceals beneath it the granite complex. This corrugated mantle, usually 
called Schists, has the granites intruding into the very base of its system 
at different periods. They cover no less than 5,000 square miles of Mysore. 
(1) The western group is the largest, occupying 3,000 square miles. It 
covers a large part of the Shimoga District. Further south it forms the 
famous (a) Bababudan belt in the Kadur District (b) Hole-Narasipur belt in 
the Hassan District and (c) Krishnarajpet belt in the Mysore District. The 
quartzites of a haematitic character, and the hornblendic rocks of the Baba- 
budans were made use of by early man, who had many settlements in 
the Kadur and Shimoga Districts. (2) The central group, which runs through 
the town of Chitaldrug has a maximum width of 25 miles. In its total 
length of 170 miles, it includes the small belts of Chikkanaya-Kanahalli and 
Nagamangala and finally disappears near Seringapattam. The quartzites 
of Talya in the Chitaldrug area have been used in the manufacture of 
palaeoliths at that site. 

(3) The eastern group, which consists of the Kolar belt of gold-bearing 


vein quartz and homblendic rocks, is only four miles wide and forty miles 

The economic geologist has made a close and careful study of the 
Dharwars because of their economic wealth. But the Dharwars had already 
provided early man with excellent raw material for his tools, namely quartz, 
haematitic quartzite, homblendic rock, and bands of chert and halleflipta. 
In the historical period, the dark grey crystalline limestone or marble, the 
emerald-green fuchsite-quartzite, serpentine and the grey potstones of this 
system were used for ornaments such as beads and carving work in the 

The granitic complex may be divided into two series, each probably 
representing a separate epoch of intrusion. The two series are called the 
older granites and the younger granites. The older granite or the Peninsular 
gneiss, the name given to a heterogeneous mixture of different types of 
granites intrusive into the schistose rocks, after the latter were folded, 
crumpled and metamorphosed, is the most widespread group of rocks in 
Mysore and South India. It consists of different types. Some ate coarse 
and porphyritic, others fine and even grained; some are dark grey to light 
grey, others pink; some are streaky, banded and gneissic and others homo- 
geneous. Broadly speaking, it includes (a) granodiorites (b) gneissic granites 
(c) banded or composite gneisses and (d) granites. 

Somewhat different from the older granites in appearance are the younger 
granites, also known as the Closepet granites." 1 2 They form a band of 20 
miles in width and run right through the middle of the State in a north 
and south direction from Sivasamudram on the Cauvery to Molakalmuru in 
the Chitaldrug District, a distance of 200 miles. Conspicuous among hills 
which belong to this series are those of Closepet, Savandurga, Sivaganga, 
Devarayana Durga, Madhugiri and Pavgada. They form a complex, which 
consists in the main of coarse granites with large platy crystals of shining 
pink or white feldspars. There are some other isolated masses forming the 
Hosdurga hills, the Arsikere and Banavar hill ranges and the Chamundi 
hill near the town of Mysore, which belong to this series. From our point 
of view, the granitic area is important because it is here that megalithic 
monuments are found most frequently. 

The charnockites, <2) which form a younger group of rocks comprising a 
series ranging from acid to ultrabasic, is not very important from our point 
of view. Syeno-diorite is an intermediate and common type. In Mysore 
the Biligirirangan range of hills belongs to this series. 

Reference must finally be made to some rocks which are very much 
younger than those discussed so far. There are numerous dykes of a 
basaltic character penetrating all the rocks mentioned. Most of them are 
dolerites, and they appear to be the “ roots ” of trap flows which once 
covered the country, as they still do in the northern part of the Deccan. 
Their distribution in northern Mysore is given in Plate II. They have 
provided most of the raw material of the neolithic celts. 

(1) Outside Mysore, the same series has been described is Fcllary gneiss. Hosur pneiss and Ralaetiat en^iss 

in South India The Dome gneiss of Bihar and the Bundelkhand gneiss also belong to the same system 

( 2 ) After the name of lob Charnock, founder of Calcutta, whose tombstone is made of this rock. The 

temples known as the “ Seven pagodas ” at Mamallapuram near Madras, arc hewn out of this rock- 


About the laterites that cap the earlier formations and that are the 
remnants of ancient soils nothing need be said here since, unlike those of 
the Madras area, they have not produced any human industries in Mysore. 

From this brief discussion of the principal rock types of Mysore the 
following sequence emerges: — 

8. Recent soils and gravels, 

7. laterite rocks capping the Archaeans in horizontal sheets, 

6. basic dykes, chiefly dolerites, 

5. youger granite (Closepet granite), 

4. chamockites, 

3. older granite (Peninsular gneiss), 

2. champion gneiss, 

1. Dharwars (schists): — 

(a) Upper Division: 

Younger conglomerates, friable quartzites, thin bands of 
limestone admixed with volcanic material. 

(b) Middle: 

Older conglomerates, quartzites, chloritic and micaceous 
schists, limestones and banded ironstones. 

(c) Lower: 

Mainly igneous, consisting of basic and acid lava. Intrusive 


Geology governs the supply and the distribution of the raw materials of 
stone-using peoples. Hence, die distribution of prehistoric sites is often 
closely related to geological features. All rocks have not the same degree of 
hardness, nor do they fracture in the same way. In northern Europe before 
the Neolithic, for example, flint was as a rule used for tools, since it could 
be easily shaped by chipping or pressure flaking. In India and South Africa 
man had to fall back on other, and generally inferior, local materials. These 
differences of raw material greatly influence the final appearance of the tools, 
as some materials lend themselves to, or even compelled man to use, special 
techniques (Burkitt, 1928, p. 26). Another instance of the influence of the 
geological element of environment can be very well illustrated by the re- 
markable though little explored megalithic culture of southern India. The 
distribution of monuments relating to that culture coincides with that of 
the Archaean rocks, gneiss and granite, perhaps because only these provided 
suitable slabs for the construction of megalithic structures. 

Later, in the middle ages, the same factor assisted in the evolution of 
the spectacular southern style of temple architecture, with its great blocks 
of stone and its pyramidal corbelled roofs. 

The following alphabetical list contains the chief types of minerals and 
rocks used by man in Mysore and encountered in archaeological sites. 

Agate: — Agate consists of a mixture of colloidal and cryptocrystalline silica 
deposited in amygdaloid cavities in the volcanic rocks known as traps. If 


banded, it is called agate. Beinj» considerably harder than the trap rock 
itself it survives well as pebbles in trap. Hence agate was much used by 
microlith-makers. Since the nodules are rarely larger than fist-size, they do 
not lend themselves to the making of lairge tools. Its homogeneity in all 
directions makes it superior even to flint from the point of view of flaking, 
but the small size of the specimens has always restricted its use. From 
the neolithic onwards, its beauty came to be appreciated and it was ground 
and polished to make beads and other ornaments. The word agate is 
actually of Indian origin (Akik) according to Wadia (1926). 

In Mysore agates are not frequent, but an agate core was found at 

Andesites: — Andesite is a grey volcanic rock containing mainly plagioclase- 
feldspar and hornblende. It has thus the composition of diorite, a term 
sometimes applied to coarse grained andesite in the south of India. It was 
often used as a raw material for celts and is a very frequent trap rock. 

Basalt: — Basalt is a blackish volcanic rock containing mainly plagioclase- 
feldspar and augite. It is found in the trap series and was much used for 
celts in the same manner as andesite. 

Bloodstone: — Bloodstone is a green variety of chalcedony speckled with red. 
It was used, where found, in the same way as other chalcedonies, but is rare. 
It is not represented in the Mysore collections available for study, though 
there is no reason why it should not occur, since it is present in the collection 
of tools in the Department of Environmental Archaeology, London University, 
from Hyderabad and Ellora. 

Camelian: — Carnelian is a translucent reddish variety of chalcedony and 
occurs in the same manner as agate (q.v.). Carnelian was used fairly exten- 
sively at Brahmagiri. 

Chalcedony: — Chalcedony is the covering name for cryptocrystalline silica 
usually containing very small amounts of water. Apart from forming in 
volcanic rocks containing cavities it occurs in veins and as nodules in lime- 
stones. It varies enormously in colour and has received many names. The 
banded varieties for instance are called agates and were used by early man for 
microlithic artifacts (see agates). 

Charnockite: — Charnockite is a collective name for a group of dark coloured 
gneisses containing hypersthene. A large mass of them occurs just to the 
south of Mysore, and small areas are encountered in the southern part of 
the State. They have never been used by early man, to the knowledge of the 
writer, and are mentioned here because die name is familiar to most workers 
in southern India. 

Chert: — Chert is a greyish variety of cryptocrystalline silica rather impure 
and varying in colour from light grey to black. It fractures flatter than quartz 
or flint but often contains joint surfaces and invisible cracks. It is avoided 
therefore where better raw materials are available. It occurs in the form of 
beds or nodules in limestone formations. Chert was used at Br ahma giri as 
evidenced by a specimen found by the writer. 

Diorite: — Diorite is a plutonic rock (like granite) but of the composition of 
andesite. The coarsely crystalline dolerites are sometimes mistakenly called 


Dolerite: — The term dolerite is much used for the somewhat coarsely 
crystalline rocks of andesitic and basaltic composition. Dolerite is used for 
the manufacture of celts. 

Granite: — Granite is an igneous rock which solidified at a considerable depth. 
It cooled slowly and is therefore coarsely crystalline. It consists of quartz, 
feldspar and mica or hornblende. As a result of weathering, granite assumes 
the shape of rounded hills and giant boulders in tropical countries. These 
boulders provided an important raw material for man from the magalithic 
period onwards. Suitable blocks were used with little or no artificial shaping. 
Once the art of stone carving had developed, giant granite boulders became 
the raw material for the remarkable monolithic architecture of South India as 
exemplified by Vijayanagar and Mamallapuram. 

Another property of granite is the tendency to platy jointing near the 
surface. Though occasionally due to magmatic phenomenon it is in India 
very commonly caused by thermal weathering. The surface layer to a 
certain depth expands in daytime owing to intense insolation whilst it cools 
at night and contracts to a similar depth. The jointing thus produced makes 
it easy to detach large plates and these have been used abundantly in the 
construction of megaliths. It is easy to understand therefore why megalithic 
sites are so often situated at the foot of a granite hill which provided the 
raw material. It was a convenient position, for all that was necessary was 
to shift the slabs and boulders down-hill. As the geological map shows 
(Plate I) granite occurs in vast areas of Mysore outside the Dharwar series. 

Gneiss: — Gneiss is a metamorphic rock of granite composition, in which some 
or all the constituents are arranged in a horizontal band. It occurs abun- 
dantly in Mysore and in many other places. It weathers after the fashion 
of granite and was then used in a similar way. 

Jasper: — Jasper is opaque cryptocrystallinc silica very like chert or flint but 
of red, brown or yellow coloration. Occasionally it is banded, as are, for 
instance, Egyptian jaspers. There are also some green varieties. Jasper 
is a very common raw material for microliths in India. The brown and red 
varieties as well as the greyish-pink ones occur at Brahmagiri. Being formed 
either in cavities like chalcedony or agate or from sedimentary cherts by 
heat alteration in contact with igneous rocks, jaspers are frequent all over 
southern India. They occur in the Dharwar series, for instance, and are 
therefore common in many rivers. 

Opal. — Opal is hydrated silica, amorphous even under the microscope. It 
occurs in much the same way as agate but has a more glassy lustre A 
milky opal, sometimes stained brown, has been used on a large scale at 
Brahmagiri by the microlith-makers. A small core in the writer’s collection 
is made from a broken pebble and confirms that this raw material was ob- 
tained from fluviatile deposits. 

Potstone: — Potstone is a magnesia-containing, ultra-basic rock of the igneous 
group. It belongs to the Dharwar system and occurs, according to Ramo Rao, 
in the areas of Hole-Narasipur and Krishnarajpet and also as lenses in the 
granulitic rocks of south-western Mysore. On the whole, it came into use 
only with the development of architecture, particularly under the Hoysalas. 

Quartzite:— Quartzite is either a sandstone in which the grains are cemented 
together with silica precipitated from watery solutions, or sandstone, the 


grains of which are fused by heat metamorphosis. The former kind is 
frequently associated with sedimentary rocks, whilst the latter is found in the 
company of schists and gneisses. The latter type is abundant in Mysore, 
though originally restricted to the Dharwar series. Being very hard, these 
quartzites form resistant pebbles which make up the large proportion of the 
river gravels. These in fact were the main source of quartzite for the mak- 
ing of palaeolithic and microlithic artifacts. Most quartzites are brown qr 
reddish brown, the colour being due to a content of haematite or limonite, 
but many other colours occur. In the microlithic of Brahmagiri, they are 
represented by the red, pink and green varieties. The green variety, though 
on the whole rare, is frequent at Brahmagiri, where it occurs in the microlithic 

Rock-crystal: — Rock-crystal is pure crystallised silica free from water. The 
crystals may be large, sometimes a foot long, and have the characteristic shape 
of a six-sided prism terminated by a pyramid. These crystals were formed 
in cavities from a watery solution, growing from the wall into the cavity. 
They are therefore associated with the amygdaloid cavities of trap rocks and 
thus occur together with chalcedonies. But they form also in open fissures 
in all sorts of crystalline rocks and even sediments. They are therefore 
widely distributed everywhere. On the whole, however, specimens large 
enough to serve as cores for the manufacture of artifacts are of local occur- 
rence. Single specimens are found on almost every microlithic site and they 
are represented both at Brahmagiri and the Bangalore sites. Occasionally an 
occurrence of rock-crystal leads to a whole industry being made from this 
material as for instance in Banda rawela, Ceylon. 

Serpentine: — Serpentine is a rock consisting mainly of hydrous magnesium 
silicate. It is formed from other rocks which contained olivine, augite or 
hornblende, either by metamorphic processes or under the influence of deep 
reaching water action. It is therefore not an original igneous rock but de- 
rived from peridotites. The colour is mainly dark green, and it has a soapy 
feel. Although it can be cut with a knife and therefore easily carved and 
polished, it is remarkably tough and not easily flaked. It has therefore been 
used by neolithic man wherever possible for the manufacture of polished 
celts. Many so-called greenstone axes are of serpentine. Serpentine occurs 
in Mysore in the same areas as potstone (q.v.), but the writer has not so 
far seen serpentine axes from Mysore localities. 

Trap: — Trap is the covering term for the vast sheets of lava which were 
poured our over the Indian peninsula in the Cretaceous and early Tertiary 
periods. They are of considerable thickness, and many sheets lie on each 
other. The north-western part of the peninsula is still largely covered with 
them and this gives the country its characteristic plateau appearance. In the 
south and east of the line from Goa to Janakpur, however, the trap has been 
eroded away, so that the underlying Dharwar series and the basement com- 
plex are exposed. There is therefore no trap lava in sheet form present in 
Mysore. Nevertheless, at some time in the Tertiary, at least in Dart, it must 
have been covered with trap, for there are numerous dykes (fissures filled 
with volcanic material) present which intersect the basement complex. These 
are the “ roots ” from which the lava sheets were once poured out. 

Petrologically, trap is of varying composition, trachytes, andesites and 
basalts being encountered. They are mostly fine grained, but the sheet lava 


is frequently porous as the result of the original presence of gas in the lava. 
For this reason, the sheet lava is not first-class material for Stone Age tools. 
On the other hand, the rocks of the dykes are extremely dense and afford 
a first class raw material. It is for this reason that the famous neolithic 
factory sites are not situated on the trap sheets of the Deccan but in places 
in the basement complex area, where trap dykes crop out. The abundance of 
celt factories in Bellary, Hyderabad and northern Mysore all belong to this 

In describing the trap dykes of Bellary, Foote (1895) included por- 
phyritic examples occurring at Kallakurti, one mile south of Malyam in the 
Rayadrug Taluk and at Hurlihal in the south-eastern part of Kudligi Taluk. 
The Harapanahalli dyke, which is nearly 8 miles long, forms hilly ridges of 
diorite. Strictly speaking the term should be confined to basic volcanic rocks, 
such as basalt. But since in practice it is difficult to find out accurately the 
composition of trap without chemical or microscopic analysis, all dense, 
greyish, greenish or blackish volcanic rocks are called by this name. The 
word trap is derived from the old Swedish “ trappa ” which means step or 

Of the 260 dykes discovered by Foote, some are of great importance. 
The Golla-Linganahalli dyke, beginning at the extreme south end of the 
Sandur Syncline crosses the Rampur spur of Mysore territory. This together 
with the Kailasa Konda dyke west of Raidurg, if continuous, would be 38 
miles in length. These dykes are conspicuous because they contain many 
high crests of jet black colour. They must have been the favoured spots of 
' neolithic ’ man of Brahmagiri, south Bellary and Rayadrug, sites which all 
lie in and round the Chinna-Hagari river. The Woddarhully dyke was also 
easily accessible to neolithic man of Mysore, being only three-quarters of a 
mile from the Jettinga Rameswara hill of the Molakalmuru Taluk, Chitaldrug 

All trap rocks have a splintery or conchoidal fracture and are devoid of 
quartz. Though hard and tough, they can easily be ground with the aid 
of quartz sand, which is harder than the minerals composing the trap rocks. 
Trachyte: — Trachyte is a white to grey volcanic rock containing mainly 
orthoclase-feldspar and a dark component, usually biotite-mica. It is found 
in the trap series and was occasionally used for celts, though no specimens 
are known from Mysore. 

Vein quartz: — Vein quartz is a rock which occurs commonly in the form of 
dykes in the igneous areas. It is very frequent in the basement complex of 
India. The crystals may be small, of sand grain size, and the vein quartz 
then resembles metamorphic quartzite. But often the crystals are large and 
more or less translucent. From the point of view of prehistoric technology, 
vein quartz was a second-rate material. Lumps of ev°n texture are small, 
and they are apt to fracture in unexpected directions. Nevertheless, being so 
very abundant this material has been used where others were scarce. The 
Bangalore industries for instance consist almost entirely of it. It fractures 
with a rough and uneven surface and the bulbs of percussion are flat. The 
colour of vein quartz is mostly white, but occasionally grey or yellow. Vein 
quartz was used by Palaeolithic man in Mysore in about 10% of the cases. 


Chapter II 



In 1861 Bruce Foote of the Geological Survey of India discovered the 
first Palaeolithic artifact in the debris of a pit in the laterite gravel at 
Pallavaram near Madras. His discovery aroused sporadic zeal among the 
few who were interested in prehistory at that time. Wynne, for example, 
discovered an agate flake near Paithan in Hyderabad, Deccan, in 1865. 
Hackett found a quartzite hand-axe embedded at Bhutra on the Narmada. 
After Foote, King and other pioneers had made their discoveries, a pro- 
longed lull followed during which Indian archaeologists were mostly engaged 
in the study of historic civilizations. 

Palaeolithic studies received a fresh impetus when in the twenties of 
this century Cammiade and Richards began further explorations in South 
India. Their vast collections were interpreted by Miles Burkitt. In the 
publication by Cammiade and Burkitt (1930) an attempt was made to 
correlate the industries with certain climatic cycles. 

In northern India the presence of Palaeolithic artifacts was first reported 
from the valley of the Soan River (now in Pakistan) by D. N. Wadia in 1928. 
Todd in 1930 noted the site of Pindigheb. Following the expedition of 
De Terra, Teilhard de Chardin and Paterson in 1935, the Palaeolithic 
industry known as the Soan was described Q939). In this publication an 
attempt was made to correlate the pluvial cycles of South India postulated 
by Burkitt with the glaciations of the North, an attempt which in the light 
of new observations made in Gujarat (Zeuner, 1950) cannot be regarded as 

In the following paragraphs, the Palaeolithic industries of India are 
briefly characterised. This is necessary in order to provide the background 
for an analysis of the finds made in Mysore. 

Punjab. The data obtained by De Terra and his co-workers indicate that 
the Palaeolithic sites of North-western India (now Pakistan) fall into three 
main groups. It is necessary to discuss these at some length, since they 
provide the framework for the Palaeolithic of India as a whole, including 

Group I of the Punjab industries consists of such sites as Adial, 
Chaomukh, Kaliar, Jammu, and Malakpur. In the boulder conglomerate 
is found an industry which has been described as ‘ Pre-Soan ’ by De Terra 
and assigned to early Middle Pleistocene, i.e. the “ second glacial stage.” 
This industry consists of very large, thick flakes which are heavily rolled 
and exhibit prominent cones of percussion, plain striking platforms at 100° 
to 125° to the long axis of the tool and frequently a large area of cortex 
on the upper surface. Secondary flaking is entirely absent with the exception 
of one specimen that was reworked at a later date. Instead of designating 
this industry as Pre-Soan, Movius (1949b) considers it desirable to call it 
the “ Punjab flake industry,” since the older term might convey a chrono- 
logical meaning which it does not necessarily possess. 


To the second group belongs the Soan industry which is characterised 
by the presence of pebble choppers, chopping tools and flake tools. The 
evidence from ten sites in the Indus and Soan valleys appears to suggest 
that the Early Soan culture belongs to the “ second interglacial stage ” of 
Northwest India. The choppers were mostly made on rounded, oval or flat 
pebbles flaked on the upper surface only, along one edge. In other cases 
pebbles appear to have been intentionally flattened on one surface to form 
the base of the tooL 

The Early Soan has three sub-groups. Sub-group A has no flakes. 
B contains Qactonian flakes and cores from which these flakes were struck. 
In the sub-group C there occur, in addition to the Clactonian, flakes with 
facetted striking platforms and tortoise cores. The implements of A are 
heavily patinated and thoroughly worn; those of B are deeply patmated but 
not worn, and those of C are fairly fresh (De Terra and Paterson, 1939). 

The Late Soan comprises two industries, A and B, of which A is earlier 
and B later. A comes from the gravel of the “ third glacial stage,” and 
B from the Potwar clay above the gravel. Though pebble tools still survive 
in A, the great majority of the tools are flakes. It looks, therefore, as if the 
industry was not a mere development from the Early Soan. It is distinctly 
Levalloisian, though the high-angled platform flakes are still prominent. 

Industry B has the same cores as Late Soan A. Fifty per cent, of the 
flakes have facetted platforms, but there is no retouch on the flake edges. 
Most of them have large primary flaking on the upper surface and have 
certain resemblances to the Late Levalloisian of Europe (De Terra and 
Paterson, 1939). 

The third group of the Punjab industries is remarkable because it 
contains the typical Abbevillio-Acheulian hand-axe complex. It is found 
at the site of Chauntra, the stratigraphical position of which, relative to the 
Soan, remains to be determined in detail on a geological basis. From the 
typological point of view the earlier material from Chauntra contains both 
Soan and Acheulian types of implements. This may be significant, indicating 
cultural fusion. The hand-axe complex recalls European and African 
affinities, whilst the pebble choppers, chopping tools and flake implements, 
which this industry has in common with Java and Burma, suggests eastern 
affinity (Movius, 1949b). 

Gujarat. Proceeding in a southerly direction, the next area where Palaeo- 
lithic finds have been studied in detail is Gujarat Geographically as well 
as typologically, it may be regarded as a kind of half-way house between the 
Punjab and South India. Sankalia (1946) described the sites and industries. 
Zeuner (1950) discussed the geological age of the industries and suggested 
that it might be approximately “ Penultimate Glaciation or perhaps a little 

From the typological point of view, the industry of the gravel phase 
(R) contains artifacts ranging from pebble-tools of pre-Abbevillian appearance, 
many of which are comparable with the Soan, to well-finished hand-axes 
of Late Acheulian type. There are also many flakes, some of Levalloisian 
appearance, and discoid cores. Among the Acheulian types, ovates and 
cleavers occur besides pointed hand-axes. This industry is not a true Soan, 
though the presence of numerous choppers indicates a relationship. Zeuner 


is inclined to regard it as a combination of Late Soan 'with Middle to Late 
Acheulian elements. 

Madras. The Palaeolithic of South India is characterised by its hand-axe 
complex. Large n umb ers of implements have been found in the terraces of 
the Korttalaiyar valley of the Chingleput District, Madras Presidency. Of 
the terraces T°, T 1 , T 2 , and T 3 which are all of Pleistocene age, T° is the 
oldest and the highest. It comprises the laterite peneplain. The three 
other terraces are cut into this peneplain, representing three stages of halting 
of the erosion process combined with some sedimentation. These halts are 
perhaps due to eustatic movements of the sea level. 

The two most important sites of the Madras ian hand-axe industry are 
Vadamadurai and Attirampakkam, studied by Krishnaswami (1938 a and b; 
1947). Early man of Attirampakkam had excellent supplies of fine quality 
quartzite from the Alicoor-Satyavedu boulder conglomerate of Jurassic age. 
Five different industries have been recognized at Attirampakkam. 

I. Industry with Abbevillian hand-axes, made on large pebbles of red 
quartzite, with thick pebble butts and showing a white cortex. They 
are identical with those from the boulder conglomerate of Vadamadurai. 
On grounds of patination Krishnaswami recognizes three stages of this 
industry. By stage three, Acheulian influence had become apparent 
and controlled flake scars are prominent. Cleavers are absent 

II. Industry comprising Early and Middle Acheulian types with laterite 
staining, on the degree of which 4 stages of the industry are distin- 
guished. The tools are much advanced over those of Industry I. 
Cleavers begin to appear. 

III. and IV. Industries with Middle and Late Acheulian forms, with little 
or no staining. 

V. With Industry V, a derived series is found. This contains rolled 
Abbevillian and Early Acheulian hand-axes. The contemporary series, 
i.e. Industry V proper, is quite fresh and free from laterite stain'ng. 
The technique shows the peak of perfection, and its Late Acheulian 
forms are comparable to those of Europe and Africa. There is an 
amazing variety of forms of hand-axes and cleavers. 

The preceding summary of the major areas of Indian Palaeolithic has 
been presented in order to provide the necessary background for the inter- 
pretation of the Palaeolithic finds that have been made in Mysore State. 
This material is still scanty; nevertheless, a general idea of its typological 
affinities can be obtained already. 


There are about ten localities at which Palaeolithic specimens have been 
found in Mysore. For the benefit of future workers, they may be briefly 
described (see Plate III). 

Tumkur District. (1) Kibbanahalli and (2) Biligere, both in the Tiptur 
Taluk. At the eastern foot of the Banasandra hill range, one mile south of 
Biligere, some Palaeolithic tools were found in a gravelly layer below a thick 
accumulation of rainwash by Sampat Iyengar (1924) who considers this as 


a Palaeolithic factory site. The occurrence of ‘hammer-stones,’ however, 
and of implements described as ‘curved saws,’ ‘arrowheads,’ ‘guillotine 
chisels,’ and ‘ circular slings,’ together with other Palaeolithic material shows 
that later material is also present on the site. 

Shimoga District (3) Nyamati, not far from the left bank of the 
Tungabhadra river, is a small town situated on the Honnali-Ayanur road in 
the Shimoga Taluk (14°9N. 75°38'E.). It occupies a favourable position, 
being in the region of transition between the Malnad with its heavy rainfall 
and evergreen vegetation and the Maidan, the open country. At this locality, 
two pebble-tools (Nos. 224 and 225) were found by Bruce Foote in a 
‘ shingle bed,’ presumably an ancient gravel terrace of the Tungabhadra, 
comparable with similar sites in the adjacent Bellary District. 

Chikmaglur District. The Chikmaglur District has two important sites and 
a minor locality. 

(4) South of the Traveller’s Bungalow at Kadur, a small town, a 
hand-axe (No. 226) was found. 

(5) From the lateritic debris of Nidaghatta, a village in the Sakrepatna 
Taluk, three implements were recovered. 

(6) Five miles south of Nidaghatta, lies Lingadahalli which appears 
to be a promising site for further work. It has yielded several implements 
now preserved in the Foote collection of the Madras Government Museum. 

Chitaldrug District. (7) Three implements have been found in the lateritic 
debris of Talya, Holalkere Taluk. 

(8) Two more come from the talus at the foot of Jyankal, another site 
in the same district. 

Hassan District. (9) Karadigudda, a minor site in the Hassan District has 
produced only one or two Palaeolithic implements so far. 

Mysore District. (10) Ranganathapura on the Cauvery (Kaveri) river in 
the Bannur Taluk appears to be a minor site. One or two implements from 
this site are preserved in the Geological Museum, Central College, Bangalore. 

Palaeolithic man’s activity in Mysore was largely influenced by the 
availability of excellent quartzite pebbles which provided the raw material for 
his tools on a majority of sites, i.e. (3), (4), (5), (6), (7) and (8). These 
sites are nor far away from rivers, and it is probable that the raw material 
was derived from terrace gravels. 


It is exceedingly difficult to decide upon the terminology to be used in 
describing the Palaeolithic implements from Mysore State. Movius (1949 
b) has introduced the terms “ chopping tool,” “ hand adze,” “ proto hand- 
axe ” and “ chopper,” but these have proved inadequate in describing artifacts 
from South India. It has therefore been considered advisable to use the 
term pebble-tool wherever the original shape of the pebble still contributes 
decisively to the shape of the implement, and the terms “ chopper ” and 
“ chopping tool,” as defined by Movius, have been retained. For the more 


developed types tie terms hand-axe, ovate, cleaver, scraper, etc., are available 
as used in the Palaeolithic terminology of Europe. The remainder of the 
characterisation has been done by the addition of descriptive adjectives. 

In the following list the artifacts are described in order of their localities 
as enumerated in the preceding paragraphs. All registration numbers given 
refer to the collection of the Madras Government Museum, unless otherwise 
stated (as for instance, those from Kibbanahalli, which refer to the Central 
College, Bangalore). The plates and figures refer to the present paper, unless 
otherwise stated. 

Kibbanahalli. There is a large collection of artifacts in the Museum of the 
Geology Department of the Central College, Bangalore, collected from this 
site by Professors Sampat Iyengar and L. Rama Rao. The artifacts were 
found in the neighbourhood of the Banasandra hill range, which is roughly 
two miles and a half from Banasandra, a small railway station of the 
Bangalore-Arsikere branch of the Southern Railway. This region covers 
an area of half a square mile. Professor Iyengar also found many imple- 
ments in the irregularly exposed gravelly layer in the nullas below an 
accumulation of rain-wash or brick-earth, which varies from 2 to 8 feet. 

The region was ideally suited for an open camp site, and the neigh- 
bouring valley ensured Palaeolithic man’s water supply. Furthermore, there 
was the availability of an inexhausibie supply of intensely-jointed quartzite, 
the raw material for his tools. The area in and around Kibbanahalli was 
surveyed geologically by Wetherell (1900-1901). The rocks fall into three 
groups, namely schists, gneiss and granite, the schists predominating. Kib- 
banahalli lies in the schist area, but quartzites occur about six miles to the 
south, not far from Banasandra, and the haematite-quartzites about 3 miles to 
the north-east. Many quartzite pebbles, however, were used, and the river 
valleys were probably more important as a source of supply than the out- 
crops. The river east of Kibbanahalli village actually comes from the 
haematite-quartzite outcrop and this is, therefore, the most probable source 
of raw material used. The implements may be classified into the following 

Hand-axes on cores : — The hand-axes represent the largest group in the 
collection and fall into three classes, namely : 

J. Rostrocarinate-like tools. 

2. A series of crude hand-axes. 

3. A more refined group, which is characterised by secondary working 
along the working edge. It comprises hand-axes as well as ovates. 

The implements which may be assigned to the first group are of two 
kinds; there are triangular pieces of rock showing smooth joint planes, 
and tapering at the end. Though there is not much working on them, they 
were probably used as grubbing tools (the so-called Eoliths of Iyengar). In 
addition, there are other specimens deliberately shaped so as to have a 
triangular cross-section. Allchin (1952) already noted that one of these 
rostros is made of quartz. 

The implements of the second group may be classed as Abbevillian in 
technique. Z6/643 is a large pebble with a few rough flakes removed from 
both sides to form a crude hand-axe. Z6/62 may be regarded as an inter- 


mediate form between an ovate and an Abbevillian hand-axe. Alternatively 
it might be classified as a “ chopping-tool,” as defined by Movius. Z6/71 
is an interesting specimen of a hand-axe, deliberately flaked on one side, 
and with natural fractures on the other. The unworked surface shows the 
beautiful bands characteristic of the quartzite in this area. Another speci- 
men worked on one face only is Z6/569, which has a pointed end and 
roundish butt. Z6/56 is a hand-axe of a pointed type. It is asymmetrical 
and may almost be regarded as a large borer, though there is no evidence 
that it has been used as such. Four large flakes have been removed to 
produce the point, and there is no zig-zag edge. Z6/631 is a well-made 
bifacial hand-axe, but it still has the zig-zag edge characteristic of the 
Abbevillian technique. It has a slightly pointed base and is almost sym- 
metrical (PI. VI, fig. 6). Z6/16 and Z6/499 are of the same type, though with 
broader bases and the former exhibits the banded nature of the quartzite. 

The third group comprises hand-axes which are comparatively thin and 
show secondary working, such as is characteristic of the Middle or Late 
Acheulian of Europe, though the raw material — quartzite instead of flint — 
has impaired the quality of the workmanship. Z6/50 retains large portions 
of the original pebble cortex, both at the butt and on the body, but has 
good secondary working along the edge (PI. VI, fig. 5). Z6/20 may be 
described as a rough ovate, rather than a hand-axe (PI. VI, fig. 4) and Z6/728 
is a slighdy more-finished ovate. Z6/510 is one of the few specimens made 
of quartz. Though slightly pointed it may be described as an intermediate 
type between an ovate and a hand-axe. It has good secondary working, and 
typologically is reminiscent of Stage V of Attirampakkam. 

Pointed flakes : — This assemblage includes a large series of flakes resembling 
hand-axes in outline. Some may have been used as such. There are 
two types; some are massive, seven to eight inches in length, struck by 
the Clactonian technique. Good examples are Nos. Z6/668 (PI. VII, fig. 6), 
Z6/128 (PI. VI, fig. 10), Z6/520, Z6/501, Z6/58 and Z6/634. The first 
two are crude massive flakes struck off from lumps of rock which were 
more or less dressed on the surface before the flake was removed. They 
show prominent bulbs of percussion and have blunt apices which would 
have been suitable as substitutes for the points of hand-axes. 

The second type consists of flakes that have been struck almost at 
right angles to the striking platform, well illustrated by Nos. Z6/601 and 
Z6/523. The former has a large bulb of percussion and the latter is 
extremely rolled. Functionally they cannot be separated from the first type. 

Some of the specimens show coarse secondary retouch on the edges, 
for instance Z6/540. Z6/450 is a specimen on a thin, triangular flake with 
good secondary working all along the edges. No. Z6/55 also belongs to 
this group. These flakes are smaller and link the group with the scrapers 
on flakes. 

Cleavers : — There is a large number of cleavers of interesting shapes, which 
may be classified according to their mode of manufacture as: 

(a) Cleavers on cores, and 

(b) Cleavers on flakes. 

Amongst the first group is an interesting example of a cleaver made on 
a pebble (No. Z6/18). It is flaked all over on both faces, except for a small 


area on one face at the butt end where a little of the pebble cortex is left. 
This specimen is transitional between a hand-axe and a cleaver. 

Nos. Z6/597 and Z6/435 (PI. VI, fig. 2) have rounded butts and almost 
straight cutting edges. They represent type L of the Attirampakkam cleavers 
as classified by Krishnaswami (1938). Z6/33 (PI. VI, fig. 7) is a fine example 
with a straight broad cutting edge and a conical butt. It has a striking 
similarity to cleavers Type J/K of Attirampakkam V. Undoubtedly it is the 
most characteristic cleaver of the whole assemblage. No. Z6/73 is made 
of quartz. 

Most of the cleavers in the second group have a straight cutting edge 
(Nos. Z6/484, Z6/612, Z6/75, Z6/648, Z6/72, Z6/32, Z6/39 and Z6/665). 
From specimen No. Z6/32 (PI. VI, fig. 3) colloidal silica has been removed 
by weathering on the bulbar face. It has a slightly pointed butt. Z6/504, 
which has a glossy black patina all over, has an oblique cutting edge. In 
No. Z6/665 the banding of the quartzite from Kibbanahalli is dearly seen. 
From the point of view of technique, No. Z6/39 (Pl. VI, fig. 1) is interesting, 
for it is asymmetrical in outline, and very roughly trimmed on both sides. 
There are, however, large patches of original cortex left on both faces. 
The cutting edge shows a notch due to the use of the implement. The 
specimen looks more primitive than the well standardized cleavers of 
Attirampakkam V. 

No. Z6/90 may be regarded as a cleaver on a thin flake. The butt 
is flat and straight, while the cutting edge is semi-circular. This is an 
unusual type. 

Beaked implements : — One of the special characteristic features of the 
Kibbanahalli assemblage is the presence of a number of beaked tools made 
both from cores and flakes. A number of flakes are removed to produce 
a point or beak by a technique generally assodated with the industry at 
Clacton-on-Sea. In the case of these Clactonian tools, the point is pro- 
duced by a ‘ three-blow-system,’ the blows being delivered in an anti- 
clockwise direction round the intended point. Two flakes are thus removed 
from one face of the core or flake, and the third flake removed from the 
other side turns the ridge between the first two flake-scars into a point. 

Many of the specimens from Kibbanahalli exhibit a striking resemblance 
to these Clactonian implements, as for instance those from the Jaywick collec- 
tion (Department of Environmental Archaeology, London University). No. 
Z6/695 (PI. VI, fig. 9) is an example. 

No. Z6/236 (PI. VI, fig. 8) is a fine tool, with a beak produced by three 
flake-scars on the upper surface and one large flake-scar on the lower. 
Z6/494 conforms to die ‘ three-blow system,’ and the point, which is broken, 
has been slightly retouched. 

When the implement is made on a flake, two blows only are often 
enough to produce the required beak. Z6/60, for instance, was made in this 
way on part of a pebble. It has retouch along one edge of the beak. 
Z6/95 and Z6/685 are small thin flakes of the same type. 

No. Z6/61 is a hand-axe-like implement on a flake, but on the bulbar 
face a secondary flake was struck oft and a sharp point formed, the up of 
which is now broken. 


No. Z6/620 is an aberrant specimen, the point being the result of a 
single blow which has detached a long flake all through the body of the 
implement. The point of impact and the bulb of percussion are present. 

Choppers : — Among die Kibbanahalli artifacts there is a group which cor- 
responds to the unifacial “ chopper ” of Movius. It consists mostly of 
massive flakes with working confined to the non-bulbar face. In some cases 
the bulb of percussion has been removed to facilitate holding in the palm. 
In this group, however, there is one specimen (Z6/611, PI. VII, fig. 5) which 
still retains its bulb and is of the rough Gactonian type. 

Chopping tools : — A number of chopping tools (in the sense of Movius) arc 
also present. They are bifacially worked and have sinuous cutting edges. 
Some of the specimens, such as Z6/76 have been worked to produce an edge 
all round, and so this group grades into that or discs. 

No. Z6/610 (PI. VII, fig. 4) is a nicely finished chopping-tool, though 
a patch of original cortex remains on both faces. It has a relatively straight 
cutting-edge. No. Z6/125 has had large flakes removed alternately from 
either side to produce a zig-zag edge. These two specimens are the most 
characteristic of the group as a whole. 

Discs : — Most of these are made on cores, but there are some flakes which 
may be classed in this group rather than as scrapers. Z6/44 is an almost 
circular disc on a core. It is weathered, probably by the removal of 
colloidal silica from the quartzite. Nos. Z6/23, Z6/551 (PI. VII, fig. 3) 
and Z6/21 are on cores. No. Z6/515 is made on a flake and is rather oval 
in shape. Nos. Z6/583 and Z6/434 still retain their bulbs of percussion. 

Flake tools : — Many smaller flake artifacts have been found in addition to 
the larger types at Kibbanahalli. Some of them are doubtless waste flakes, 
but a large percentage have been retouched and used. It is difficult to 
classify them functionally since each group grades into the next, but they 
can be roughly divided as follows: — 

Nos. Z6/471 (PI. VI, fig. 11), Z6/107, Z6/544 and Z6/555 (PI. VII, 
fig. 2) may be regarded as points. Z6/544 retains some cortex, but one 
end has been worked. Retouch can be clearly seen running along one edge 
of the point. Z6/521 is a Gactonian flake with the bulb of percussion 
still present. Three small flakes have been removed from the bulbar face 
to form the point. Z6/646 is a well-finished specimen. It is a thin flake 
which has had a small secondary flake removed from the bulbar face and 
steep retouch along both edges of the point. Nos. Z6/104 and Z6/714 are 
triangular flakes which may be regarded as points; the latter shows marks 
of use along one of the sides of the point. No. Z6/448 has been described as 
a Mousterian lance-head by Sripada Rao (1932). There is, however, noth- 
ing typical of the Mousterian in the artifact. It appears to be a plain 
“ Levalloisian flake,” 4 ins. long with an unprepared right-angle platform. 
It shows slight retouch which has been worn, probably by rolling. It may 
be regarded as a point. 

No. Z6/449 is an asymmetrically pointed flake of stained fuchsite-quart- 
zite, 4.3 ins. in length. There is some working on the flake but hardly 
any retouch, and certainly no resolved flaking (Sripada Rao, 1932). It 
also appears to be a simple “ Levalloisian flake ” with an unprepared right- 
angle platform. Functionally a point. 


There is still another group of retouched flakes which may be regarded 
as borers. The best example of this group is No. Z6/593 (PL VII, fig.l) 
which is a thin flake, about 3 ins. long, with a steep retouch along one side 
of the point. This retouch produced a concavity which emphasises the 
sharpness of the point and would have facilitated boring holes in skins 
and hides. 

Scrapers: — All types of scrapers are present in the collection and they are 
mostly made on flakes. Side scrapers and hollow scrapers dominate the 
whole assemblage. 

End scrapers: — They are rare, but two examples may be noted. Z6/82 
is a thick, almost circular flake, but as it has been retouched along about 
2 ins. of the periphery to make a straight edge, it may be classed as an 
end scraper. No. Z6/42, a thin flake, has the end retouched only on the 
bulbar face. No. Z6/88 might be regarded as a rather coarse end scraper. 
There were some tools which were used as side scrapers or end scrapers 
as occasion demanded, such as Z6/103 (PI. VII, flg. 8) and Z6/486. 

Side scrapers: — The flakes which may be put in this category are those 
which have little or no retouch, but appear to have been used as scraners 
along one or both sides (Nos. Z6/106, Z6/456, Z6/96 and Z6/524). This 
category grades into the concave or hollow group, since re-sharpening 
or use is apt to produce a concavity. Many of the side scrapers are, 
in fact, composite tools, for example Z6/237 which has a notch or hollow 
in addition to the straight cutting edge. 

Hollow scrapers: — Hollow scrapers form a large pan of the collection and, 
as in the Clactonian industry, they indicate that the working of wood was 
practised by early man of Kibbanahalli, though no artifacts of wood have 
survived here. 

In this class of scraper, a large flake has been removed to produce 
a concavity and then steep retouch within the concavity has strengthened 
the edge (No. Z6/93, PI. VII, fig. 7 and Z6/586). In some cases, such 
as Z6/683 and Z6/433, the retouch appears to be on the bulbar face, 
although the flake which produced the concavity was taken from the upper 
side of the scraper. In other cases, the splitting of a pebble by thermal 
fracture has left a slight concavity and this has been retouched (Z6/433). 
Another type within this group is the double-hollow scraper (No. Z6/126). 

Fabricator : — A fabricator is a tool used for producing secondary working 
or retouch on stone artifacts. It is often a long, pointed flake convenient 
fox retouching the cutting edges of hand-axes. The specimens placed 
in this category are rough, and therefore somewhat doubtful. No. Z6/48 
(PI. VII, fig. 9) shows what appears to be bruises caused during the opera- 
tion of pounding the rough-outs. No. Z6/666 also has the bruises, but 
is slightly rolled. 

Cores: — Kibbanahalli has yielded some remarkable cores. The ordinary 
“Clactonian” cores from which the large flakes were struck are not in 
the collection, but they must have been part of the industry. The aberrant 
cores suggest techniques of striking blades which would be surpr ising in 
a Lower Palaeolithic context (Z6/122, Z6/133; PI. VII, figs. 11, 10). They 
are large editions of the cores common in much later microlithic industries. 


The blades struck from them, however, are not represented in the collection. 
One is, therefore, left with the problem whether blades are part and 
parcel of the Kibbanahalli industry or whether there is a later blade 
industry to be found at that locality in addition to the Lower Palaeolithic 
one, of which the cores only have so far been found. 

The most interesting core is Z6/122 which is a conical blade-flake 
core. It is a proto-type of the conical blade-cores of microlithic industries. 
It is about 5 ins. long and has had six or seven flakes removed. Z6/503 
is another, from which four flakes have been removed. 

That some of the large, thick flakes were also used as cores at Kibbana- 
halli is exemplified by Z6/2. It is a thick flake, about 7 ins. long, from 
which a flake of about 4 ins. long and 3 ins. broad has been detached. 

Nyamati. No. 224 (PI. XI, fig. 7). Bifacial asymmetrical cleaver-like pebble 
tool with rounded working edge. Working edge approximately semi- 
circular, the opposite end being formed by unworked part of the pebble 
which includes a natural flat base on which it is possible to stand the 
specimen upright. L. — about 7 cm. W. — about 8 cm. Material quartzite. 

No. 225. (PL IX, fig. 11). Bifacial pebble tool with a pointed working 
edge. Flaking appears to be free with very little secondary working. A 
large part of the original pebble, which is flattish, is unworked on the 
underside of the implement. Functionally, the specimen may be regarded 
as a hand-axe. L. — about 9 cm. W. — about 8 cm. Material — quartzite. 

Kadur. No. 226. (PI. IX, fig. 5). Pointed hand-axe, worked bifacially. 
Portion of the original pebble-surface left unworked on the butt portion. 
Typologically Abbevillian, though such specimens are known to occur in 
India up to the Late Acheulian. L. — about 8 cm. W. — about 6 cm. 
Material — quartzite. 

Nidaghatta. No. 209. (PI. X, fig. 4). Bifacially worked disc. Coarse 
free flaking all round, about six flakes having been taken off on each side. 
Cutting edges fairly straight. Some part of the original pebble surface 
is left in the middle of die sides. L. — about 9 cm. W. — about 8 cm. 
Material — quartzite. 

No. 210. (PI. XI, fig. 1). Small polyhedric disc. Worked on both 
sides. Bold free alternate flaking, five or six flakes having been taken off 
on each side. Cutting edge zig-zag. L. — about 7 cm. iW. — about 6 cm. 
Material — quartzite. 

No. 211. (PI. VIII, fig. 1). Beautifully-made symmetrical hand-axe 
with a rounded apex and a slighdy zig-zag cutting edge. Displays good 
secondary step-flaking. Retains a small portion of the original pebble 
surface on both sides. Has a flat butt due to an original cleavage plane 
of the rock. L. — 7.5 cm. W, — 6 cm. Material — quartzite. 

Lingadahalli. No. 212. (PI. X, fig. 7). Pointed hand-axe. Like 211, has 
a flat butt and displays zig-zag cutting edge produced by bold alternate 
flaking. Tip broken off. Typologically, it might be classed as Abbevillian, 
but Acheulian rough-outs are of the same kind. L. — 8 cm. W. — about 
6 cm. Material — quartzite. 

No. 213. (PI. VIII, fig. 8). Subovate hand-axe, probably worked 
bifacially. Difficult to make out the flake scars since the material is white 
quartz. Cleavage planes of the rock have been taken advantage of in 


shaping this specimen. The butt end is more or less straight. L. — about 
8 cm. W. — 6.5 cm. 

No. 214. (PI. IX, fig. 9). Small asymmetrical hand-axe, worked on 
both faces. Bold alternate flaking has produced the zig-zag cutting edge. 
The butt retains considerable portion of the original pebble surface. 
Typologically, this specimen would be classified as Abbevillian in Europe. 
L. — 6.5 cm. W. — 6 cm. Material — quartzite. 

No. 215. (PI. VIII, fig. 9). Irregular ovate, bifacially worked. One 
end more flattened than die other, which may be regarded as the butt. L. — 
11 cm. W. — about 8 cm. Material — white vein-quartz. 

No. 216. (PI. VIII, fig. 10). Finely pointed hand-axe made from a 
thick flake. Some secondary work (step flaking) is present after free flak- 
ing was used in order to straighten the edges. This suggests an Acheulian 
rather than Abbevillian technique. The two sides are different, one being 
made up in part of the primary flake surface. But at right angles to the 
bifacial plane, this implement is almost symmetrical. Flaking coarse, but 
the edges are straight. L. — about 8 cm. W. — 5.5 cm. Material — quartzite. 

No. 217. (PI. VIII, fig. 3). Like No. 216 it is a pointed bifacially worked 
hand-axe though it is pronouncedly asymmetrical. Part of the original sur- 
face is still present. Free flaking with zig-zag edge. Secondary step- 
flaking negligible. Butt end wedge-like, almost pointed. L. — about 9 cm. 
W. — about 5 cm. Material — quartz. 

No. 218. (PI. X, fig. 9). Several large flakes were taken off to make 
the upper face of this tool. Since the underface is plain, except for two 
smaller flake scars, the implement may be classified as a chopper according 
to the definition of Movius (1949b). Owing to rolling, secondary working 
is not discernible, but it is clear that the tool was made from a heavy 
flake. The flake was a Clactonian one, using a pre-existing joint surface 
as striking platform. This specimen shows that “ choppers ” are not always 
core tools. L. — 8.5 cm. W. — 6.5 cm. Material — quartzite. 

No. 219. (PI. VIII, fig. 4). Suboval hand-axe made on a pebble. The 
under face is almost flat. Point produced by rough bilateral flaking. About 
half the implement still covered with original pebble surface. Somewhat 
worn. This specimen is of the “ Abbevillian ” type. L. — 7 cm. W. — 
4 cm. Material — quartzite. 

No. 220. (PI. VIII, fig. 5). Small hand-axe on a flat flake with rounded 
point. That the flake was struck at a high angle, Clactonian fashion, is 
shown by the cone of percussion. This specimen may alternatively be 
regarded as an abnormal cleaver, for the following reasons. Its cross- 
section is a parallelogram, the flaking of both upper and undersides being 
confined to the left margin. The lower edge (presumed cutting edge of 
the cleaver) is flaked on the upper side only. It is, however, not so straight 
as it should be in a cleaver. The upper end is flattened by one flake scar, 
on the upper side, giving it a duck-bill shape. Perhaps this was a dual- 
purpose tool which could be used as a cleaver or as a hand-axe. L. — about 
7 cm. W. — 5 cm. Material — quartzite. 

No. 221. (PI. X, fig. 11). Polyhedric ovate, with zig-zag cutting edge. 
Upper side pyramidally raised. There is a short, cutting-edge produced 
by a flake scar on the upper side and a joint surface on the lower. It is 


a core tool and resembles the “ chopping-tools ” of Movius. L. — 6.5 cm. 
W. — 5 cm Material — quartzite. 

No. 222. (PI. IX, fig. 1). Disc-like “ chopping-tool.” Sub-circulai 
bifacially worked, but with a “ butt ” portion left flat. Step-flaking present 
but probably due to the effect of the bedding-planes. Hence there is no 
evidence of the step-flaking being intentional. L. — 8.5 cm. W. — 7.5 cm. 
Material — banded quartzite. 

No. 223. (PI. XI, fig. 5). Fine asy mm etrical ovate, made on a flake. 
It has a long striking platform of 5 cm. and most of the underside consists 
of the flake surface. That the flake was struck in the Levalloisian fashion 
is shown by the right angle position of the striking platform to the axis 
of the implement. The bulb of percussion is still present. The edge 
opposite the striking platform is bifacially flaked to form a zig-zag edge. 
Functionally a chopper or a cleaver, but cutting-edge irregular and not 
continued all the way round. Whether the tip is significant cannot be 
decided. L. — 7.5 cm. W. — 6 cm. Material — quartzite. 

Jyankal. No. 207. (PI. VIII, fig. 2). Pointed asymmetrical hand-axe. 
Shape greatly influenced by cleavage of the rock. Edges flaked from both 
sides. Tip rounded. L.— -8 cm. W. — about 6 cm. Material — quartzite. 

No. 208. (PI. VIII, fig. 6). Semi-circular “ chopper,” on coarse 
flake. Cutting edge worked only on one face. Butt-end a straight jointing 
plane of the rock. L. — about 8 cm. W. — about 9 cm. Material — banded 

Jodi Katte. No. 203. (PI. XI, fig. 3). Asymmetrical short duck-bill hand- 
axe of sub-oval shape, made on a flake. Shows flaking round the edge, 
mainly on upper side. L. — about 7 cm. W. — about 6 cm. Material — 

Talya. No. 204. (PI. IX, fig. 3). Round-pointed asymmetrical hand-axe. 
The bold alternate flaking has produced a sinuous edge. Large portions of 
original surface left on both sides, possibly made from a pebble. It is 
noteworthy that the butt-end has been made narrow by flaking from both 
sides. L. — about 9 cm. W. — about 6 cm. Material — quartzite. 

No. 205. (PI. X, fig. 1). Ovate of fine workmanship. Nearly sym- 
metrical. Although the implement is worked all over on both sides, one 
face retains a portion of the original surface. On the opposite side, a large 
scar suggests that part of the surface was lost owing to a natural accident 
after completion of the tool. This is possibly an insolation fracture. 
L. — about 11 cm. W. — 8.5 cm. Material — quartzite. 

No. 206. (PI. IX, fig. 7). Irregular. Difficult to study the working 
on the implement owing to the bedding planes producing the effect of step- 

flaking. Worked on both faces all round. L. — about 8.5 cm. W. 

6.5 cm. Material-banded quartzite, same as No. 222 from Lingadahalli. 
Karadigudda. No. 202. (PI. VIII, fig. 7). Irregular ovate, worked all the 
way round on both sides. The upper right margin of the upper side has a 
concave “ bay ” produced by a flake struck from the underside. It cannot 
be decided whether this was intentional or is an accident. L.— about 8 cm. 
W. — about 5.5 cm. Material — white vein-quartz. 

Ranganathpur. Z6/441. Geological Museum, Central College, Bangalore. 
Appears to be a rough hand-axe. It has been considered as a new type 


(Sripada Rao, 1930) because it is supposed to contain two small ledge-like 
notches suitable for hafting. In a poor raw material like quartz, it is not 
admissible to consider these notches as intentional. This specimen resembles 
No. 213 of the Madras Museum. L.— about 9 cm. W. — about 7 cm. 
Thickness — about 3.3 cm. 



The typological position of the Mysore Palaeolithic industries cannot 
at present be fully assessed. Kibbanahalli has produced a larger collection 
of artifacts than any other Mysore site so far investigated. But it is not 
the magnitude of the site as compared with other Mysore sites which affects 
the problem of correlation, so much as the complicated nature of the industry 
itself. Together with the usual assemblage of South Indian Palaeolithic 
artifacts, there occur at Kibbanahalli a number of beaked tools and Clactonian 
types, which are apparently alien to the Lower Palaeolithic industries of 
neighbouring regions. Further investigation of these regions, which might 
reveal comparable industries elsewhere, is required before the question of the 
position of Kibbanahalli, and therefore also of Mysore itself can be finally 

Apart from the aspects of the Kibbanahalli industry mentioned above, it 
is possible to treat the Mysore material as a unit, although it comes from a 
number of different sites. This is due to its remarkably uniform character, 
expressed mainly by primitive shapes and the rarity of Late Acheulian speci- 
mens. The number of artifacts (except those from Kibbanahalli) so far 
collected is too small to give a final verdict. Considering, however, that 
occasional collecting usually produces the more elaborate types of tools, 
simply because they are more conspicuous and more easily found, it is 
justifiable to say that a late Lower Palaeolithic industry appears to be 
lacking in Mysore. This statement can be substantiated by a comparison 
of the Mysore specimens with the industries found at Attirampakkam, 
Chingleput District, Madras Province. These industries have been studied 
by Krishnaswami (1938a). The sites were visited by the writer with 
Professor Zeuner’s expedition in 1949, and there is a clear geological sequence 
available into which successively more evolved industries have been fitted. 
The material has been studied in an endeavour to determine the typological 
stage to which the Mysore material might point. The results of these com- 
parative studies are given in the following paragraphs and illustrated in the 

Before the present attempt was made, Sankalia (1946) had already drawn 
some parallels between his Gujarat material and specimens from Mysore. 

He had compared: 

Sankalia’s No. 

No number 

Pedhamli III 
Pedhamli III 
Pedhamli II 

Foote’s No. 
with 202 
with 205 

with 216 
with 209 








Sankalia’s comparison would indicate that specimens comparable with 
the Acheulian industry of Gujarat occur in Mysore. The Gujarat industry 
(see page 13) belongs to the Acheulian complex. In view of its Soan 
affinities however, it is advisable from the present point of view to confine 
the comparison to the nearest available sequence of Palaeolithic industries, 
that of Attirampakkam. 

The Mysore Palaeolithic comprises tools made from pebbles, as well 
as pieces of solid rock. Quartzite pebbles being abundant in the rivers of 
Mysore, it is not surprising that pebble tools should occur in this State as they 
do in Gujarat and Attirampakkam. An interesting feature is the presence 
of specimens made from vein quartz, a material which is readily obtainable 
from the quartz dykes of the Basement Complex of Mysore. Owing to the 
bad fracturing qualities of this material, however, tools made from it are 
rarely informative from the typological point of view. Another raw material 
used, a banded quartzite with a tendency to split along the bedding planes, 
is not known from elsewhere. It is probably derived from the Dharwar 

The pebble tools found in Mysore are not of the primitive type illus- 
trated by the specimens from Oldoway in East Africa. The Mysore speci- 
mens are flaked in a more or less complex manner and are bifacial. 
Typologically, therefore, they may be regarded as specimens of an Abbe- 
villian kind of workmanship (excepting No. Z6/50 from Kibbanahalli), 
although no chronological correlation with the Abbevillian of Europe is 
implied. In fact, pebble tools of the Oldowan type are quite frequent in the 
Acheulian of the Sabarmati. Neither the technological stage nor the age of 
the Mysore Palaeolithic, therefore, is indicated by their presence. In the 
material from Attirampakkam III and from Chintaladevi (24 miles north- 
west of Kavali, Nellore District) pebble artifacts occur which are very similar 
to those of the Mysore area (see PI. IX, figs. 11 and 12). 

A few rostrocarinate-like implements from Kibbanahalli may point to 
some pre- Abbevillian elements in the Palaeolithic industries of Mysore, 
although sometimes an unfinished tool may have the appearance of this type 
of implement. The problem is worthy of further investigation, however. 

The pebble tools grade into more or less pointed hand-axes of the 
ordinary type. Most of these display the zig-zag edge of the Abbevillian 
technique, and step-flaking deliberately applied to straighten the edge is 
present only exceptionally (No. 211, No. Z6/510). Of the well-worked hand- 
axes, Nos. 216 and 217 bear a striking similarity to Attirampakkam III, 
51 and AV III, 5 from Vadamambakkam near Arkonam, Madras Province 
(PI. VIII, figs. 11 and 12). No. 204 from Talya resembles AT III, 38 
(PI. IX, figs. 3 and 4). 

The hand-axes in turn grade into mqxe- or less oval, bifacial tools, 
a few of which may legitimately be callefkbvates ‘(Nos. ,Z6/728, Z6/20 and 
205). The best specimen is No. 205, Jfyieh is comparable with European 
Acheulian ovates. Typologically, it is/'fcharacterised by flaking which pro- 
duces a straight, untwisted edge and which leaves one side thicker than the 
other. Altogether the specimen is thidk by European standards, but it must 
not be forgotten that most of the raV material used in Europe is flint, 
whilst in India it is mostly quartzite, whHh' fs a jechnically difficult material 
to work. The straightening of the edge appeals, to have been achieved by 


a certain amount of step-flaking which, for the same reason, is somewhat 
coarse. In the European classification this specimen would pass for Early 
or Earlv-Middle Acheulian. Similar specimens occur in large quantities in 
the Middle Acheulian of East and South Africa. In south India comparable 
specimens are found at Vadamambakkam (AV III, 9) and Uliyambakkam 
(AV VII, 3, see PI. X, fig. 2). 

At Attirampakkam similar specimens occur in phase V, but apparently 
not in phase III. No. AT V 65 has been illustrated in PI. X, fig. 3. Finally, 
specimen No. 205 is comparable with the ovate that Cammiade (see Richards, 
Cammiade and Burkitt, 1932) found on the left bank of a stream to the west 
of the Polavaram Taluk in the Godavari District. The ovate from Talya is, 
from the technological point of view, the most advanced palaeolith so far 
found in Mysore. In addition to ovates, choppers of the Movius type occur. 
No. Z6/611 is a fine example. Discs are too numerous to mention. They 
are of widespread occurrence in India, although they tend to be more frequent 
in the earlier phases of the hand-axe culture. Of the Mysore specimens, 
Nos. 206, 209 and 222 can be matched with specimens from Attirampakkam 
II (AT II, 88, PI. IX, fig. 2) and from Kilkuppam near Arkonam (AV II 17, 
PI. X, fig. 5). 

The choppers pass into types of tools which have chopping edges only 
round a portion of their circumference, leaving at the butt-end a flat “ base.” 
This base, however, is usually due to advantage being taken of the presence 
of a cleavage plane in the rock. Of this group, Nos. 211, 212, 223 and 224 
may be mentioned. Similar specimens occur in Attirampakkam III; Nos. 212 
and 223, for instance, are comparable with AT III 54 and AT III 191 (PI. X, 
fig. 8 and PI. XI, fig. 6) respectively. From the functional point of view these 
specimens may be regarded as simple cleavers. They certainly have not 
the characteristic shape of the cleaver as described by Foote. 

There are many true cleavers, particularly from the Kibbanahalli site. 
They are of many types and Nos. Z6/33, Z6/597 and Z6/504 are carefully 
finished and may belong to the Attirampakkam V stage. Some of them are 
earlier and belong possibly to Attirampakkam III. Nos. Z6/665 and 
Z6/39 show considerable economy of work. No. 220 is another interesting 
specimen; with its parallelogram cross-section and its tapering outline it is 
reminiscent of type AB of Krishnaswami (1938a). 

To sum up; it is apparent that the Mysore Palaeolithic (as far as the 
available material suggests at the moment) represents a culture comparable 
with Attirampakkam III. This culture indeed contains all the types that have 
so far been found in Mysore. The only exceptions are ovates No. 205 and 
a few others from Kibbanahalli which might be later. 

Kibbanahalli presents certain special characteristics. It includes some 
very primitive material as well as some specimens of late appearance. Its 
beaked tools together with the prolific occurrence of the hollow scraper 
indicate a strong affinity with the Clactonian. This aspect of the Kibbana- 
halli Palaeolithic industry stands alone and finds no parallel in the rest of 

Of course, the Mysore material is not all of one age. Nevertheless 
the bulk suggests that a culture of the Attirampakkam III type was relatively 
widespread in Mysore State. More finds will undoubtedly be made in the 
future and extend the typological range of the Mysore Palaeolithic. 


Chapter III 



It is a curious fact that in the whole of South Asia a true Upper 
Palaeolithic in the sense of the European sequence has not yet been recovered. 
In view of the fact that numerous sites have become known both of Lower 
Palaeolithic and microlithic industries, the least that can be said is that such 
Upper Palaeolithic industries must be very rare. 

It must be admitted that in the late Soan of N.W. India (now Pakistan) 
a flake element becomes rather prominent, so that it is possible that the 
hand-axe industries were followed by a period of flake industries based on 
the Levalloisian technique. But sites with pure and typical industries are 
still wanting. On the other hand, the oldest industry found by Subba 
Rao (1948) at Sanganakallu, Bellary District, appears to contain a Levalloisian 
element. These patinated flakes of trap and sandstone, of which about 400 
have been found, are at least in part made from cores prepared in the 
Levalloisian fashion. This is particularly evident in the case of the short 
round flakes, which on the upper surfaces clearly show a preparation by 
flaking from the periphery of the core, prior to the taking off of the flake 
which it was intended to strike. Similarly the long flakes are struck from 
cores from which other long flakes had been taken off in the same direction. 
This might be a variety of Levalloisian technique, as indeed it is wherever 
the so-called blade-flakes were made by peoples using the prepared-core 
technique, for instance in Mark Kleeberg (Central Germany), on the Somme 
(Northern France), in the Mount Carmel caves (Palestine) and elsewhere. 
Where, however, the raw material is not first-class flint or obsidian, poor, 
flake-like blades are easily produced from cylindrical or conical blade cores. 
This possibility applies to the long flakes from Sanganakallu since they are 
made of trap and sandstone. If this is the case, they are no more than a 
large facies of the microlithic blades present in the same layer. This alterna- 
tive has to be regarded seriously, since Subba Rao expressly states that 
microliths were actually found in association with the flakes under discussion. 

A similar occurrence of blade-flakes in association with a microlithic 
industry is observed also in a collection of implements from Hyderabad made 
by von Furer-Haimendorf and housed at the Institute of Archaelogy, London. 
It is composed of surface finds and contains large numbers of rough fluted 
cores which may be regarded as giant forms of the microlithic cores. With 
them occur numerous flakes comparable with the Sanganakallu ones and 
also many microliths. The raw material is jasper and chert, available in 
large lumps. The age of this industry is unknown; it may have lasted into 
comparatively recent times. 

It is noteworthy that blade-flakes are present at Brahmagiri also, but 
their association with microliths cannot be established since they have so far 
been found on the surface only. This will be described on page 34. 


This evidence is admittedly scanty, though it leaves one with the impres- 
sion that the microlithic industries of South India can be derived from a 
hypothetical Levalloisian flake industry which preceded them. 

In this respect, South India would not stand alone, since in North 
Africa, in the opinion of Professor Vaufrey (1933), the Aterian, a direct 
descendant of the Levalloisian, passes into microlithic industries such as 
the Oranian (Ibero-Maurusian). The case is similar in East Africa, where 
the Levalloisian Stillbay, in turn, passes into the Magosian and Wilton B 
(Leakey, 1936). 

The only element in the South Indian Stone Age which might con- 
ceivably be regarded as Upper Palaeolithic is the ‘ Industry III ’ of the series 
of Cammiade and Burkitt (1930) from the Nandikanama Pass and other 
localities. It is characterised by the presence of “ slender blades with blunted 
backs, a few burins, planing tools and end-scrapers.” 

It must however be noted that these elements occur in the microlithic 
industries also. Moreover, the collection figured by Cammiade and Burkitt 
on their plates V and VI would not be out of place in the Mysore microlithic 
industries of the Jalahalli group. The fluted cores, crescents and points are 
particularly striking. So far as evidence goes at the moment, therefore, there 
is no need to assume the presence of an independent Upper Palaeolithic in 
Mysore or elsewhere in South India. 

Until definite evidence to the contrary is brought forward, the microlithic 
period must be regarded as an immediate successor to the vast period of 
industries of Lower Palaeolithic aspect. 


It is worthwhile noting, as briefly as possible, the several assemblages 
of microliths discovered in India. It was in 1867-68 that Carlyle (1) bepan 
collecting microliths in the caves and rock-shelters of Bundelkhand, Baghelk- 
hand, Rewa and parts of the Kaimur range. Since his discoveries, a large 
number of other sites have come to light. They have been listed by D. H. 
Gordon (1938). To this list must be added the Gujarat sites discovered 
and systematically excavated by Sankalia (1944, 1945 and 1946) and the 
Godavari sites of Cammiade (1924), more than a hundred in number. Maski 
in the Raichur District of Hyderabad State, Sawyerpuram, Kulattur and 
other sites of Tinnevelly District and Todd’s (1939) sites round Bombay are 
the most important found in recent years. Mention must also be made 
of the Kurnool sites of Cammiade and Burkitt and Patpad of Foote in the 
same area. Codrington’s site of Ellora and the Jabbalpur sites of Gordon 
are of potential importance since they have produced rare types of implements 
which I have been able to examine in die collection of the Institute of 
Archaeology, London. 

The recorded sites thus extend from Karachi (Pakistan) on the west to 
Seraikala in Bihar in the east and from Jamalgarhi (Pakistan) in the north 
to Tinnevelly in the south. 

(I) Specimens of his collection are found m the Hritiah Museum, London, and Pitt-rivers Museum, Oxford, 
where I have been able to examine them, and in the National Museum, Dublin V. A. Smith <1906) 
published a short paper on them. 



A systematic search for microlithic sites in Mysore State is likely to lead 
to the discovery of many more localities than are known at present. As early 
as 1895 Bruce Foote, who then was Mysore State geologist had discovered 
in many places cores of chert and chalcedony together with some worked 
flakes. French Rocks near Seringapatam yielded a few cores. Near the 
Travellers’ Bungalow at Hiriyur, Chitaldrug District three flakes were picked 
up by chance. A worked flake of dark chert came from Kaldurgahill in the 
Tarikere Taluk. A thick, triangular flake of chert and a core of Lydian stone 
were found at Holakal Hill in the Sira Taluk by Sambasiva Iyer of the Geo- 
logical Survey of Mysore. Kinciad Lee found a few flakes and cores on the 
race course at Bangalore, and in 1896 he found a number of flakes and cores 
on Minakshayya’s estate, south-east of Tumkur. Recently D. H. Gordon 
(1945) discovered two sites near Bangalore, Apart from finished implements 
which are rare, a considerable quantity of cores and flakes has been recovered 
from these sites.' 1 11 There is no doubt, therefore, that they are manufacturing 
sites. One of the sites is on a bare piece of rising ground to the North of 
the road which leads to the Hindustan Aircraft factory at a point five and a 
half miles outside Bangalore. The second site lies about a mile and a half 
to the south of the Sarjapur road between the ninth and tenth milestones 
from Bangalore. This locality also is a low bare hill which dominates the 
whole neighbourhood. Owing to the refractory nature of quartz, the sole 
raw material of these two sites, there is, of course, a high percentage of 
failures. A few flakes also occur on the surface at Siddapura and Jeting- 
arameshwara, which are in the neighbourhood of Brahmagiri. 

Apart from these stray discoveries, a few sites in Mysore State (PI. Ill) 
have produced large numbers of microliths. One of these is Jalahalli, about 
ten miles north-west of Bangalore and one and a half miles from Yesvantpur 
Railway Station on the metre-gauge line to Poona; another is Brahmagiri, in 
the Chitaldrug District, famous for its Asokan rock edict; and the third is 
Kibbanahalli, Tumkur District. 

At Jalahalli, close to 137 Indo-British General Hospital compound, two 
wedge-shaped granite hils are seen; the nearer hill rising to the west, the 
other, rising to the south. Between the two hills runs a narrow valley 
about a quarter of a mile wide. According to Todd, who described this 
site (1948, p. 27), the granite hills in places contain intrusive veins of milky 
quartz and rock crystal. 

On the surface of the nearer hill Mr. Todd discovered in November, 
1946, flakes and implements of microlithic type. The farther hill, which 
was being quarried for granite to be chiselled into slabs, had had the weather- 
ing soil cut away and removed. Below this black soil, there is in places 
a reddish soil horizon described by Todd as “ pellety laterite.” If it were 
a true laterite, the contained microliths, being made of quartz, would have 
perished. It is evident, therefore, that this horizon is of a colluvial nature, 
containing “ lateritic ” pellets formed elsewhere and mixed with decomposed 
fragments of local granite. The microlithic industry found here is similar 

(1) I hr implements from these two sites and also those from Dhirwar. Rtlgaum, Chandargi Piranwadi, 
jahb.tlpur, Mohcnjo-daro and Dighi Jiill ncjr Poona, forming his collection are housed at the Institute 

of Archaeology Grateful thanks are due to Col Gordon for valuable information about these sites 


to the one found on the nearer hill. The implementiferous horizon is from 
18 inches to 5 feet below the surface and in all cases lies either on the granite 
(the parent rock) or on the “ lateritic ” deposit. 

The implements therefore occur sometimes on and sometimes in the 
red layer, which consists of a rubble of decomposed granite fragments 
mixed with limonite pellets and is fairly firmly cemented. Some of the 
microliths had dropped into the large cracks and fissures of the parent 
rock, where they are to be found. 

As regards raw material, it is interesting to note that all the implements 
were exclusively made from quartz and rock-crystal. In this respect, they 
closely resemble the artifacts from Calient in the Madras Presidency and 
Bandarewela in Ceylon, which I have been able to study both in the British 
Museum and in Oxford. Out of the whole Todd Collection of implements 
from Jalahalli now in the British Museum, only one is of red jasper. Quartz 
being a difficult raw material, the evidence for the flaking and the retouch 
on the implements is not easily detectable, but those made of rock-crystal 
display beautiful workmanship. 

Brahmagiri is a granite outcrop rising some 600 ft. above the plain in 
the Molakalmuru Taluk of the Chitaldrug District in the northern extremity 
of Mysore State. It sprang into great importance in Indian archaeology as 
a result of the discovery of three copies of Asoka’s rock-edict No. 1. This 
find led to the conclusion that the northern portion of Mysore formed the 
most southerly known point of the great empire of the Mauryas. The men- 
tion of the town of ‘ Isila ’ in the rock-edict, led to a deliberate search. The 
town-site was eventually identified, lying at the foot of the Brahmagiri hills. 
With a view to uncovering the Mauryan town, trial excavations were carried 
out by Dr. Krishna of the Mysore Archaeological Survey in 1940 and 1942 
and, the great potentialities of the site being amply confirmed, careful exca- 
vations were undertaken by Dr. Wheeler on behalf of the Government of 
India in 1947. 

The northern slopes of Brahmagiri are largely covered by a tumbled 
mass of granite boulders. There are extensive signs of ancient occupation 
in the form of potsherds, fragmentary walls and remains of small terraced 
platforms roughly revetted by dry-stone walling. It is to be presumed that 
2000 years ago there was more earth amongst the boulders than remains at 
present and that much evidence of this part of 1 Isila ’ has been washed away. 
The main area of occupation must have lain, as surface remains and exca- 
vation combine to indicate, along the gentle slope which now forms the 
transition from the hill to the plain. In the midst of the zone of occupation 
stands the great boulder which bears on its upper surface the best-preserved 
of the three copies of the edict. A furlong to the south-east, up the hill- 
side, there is a small brick-chaitya, a sadly ruined structure but nevertheless 
the best preserved monument of Isila. 

In order to fit the microlithic industries into this picture of the site of 
Brahmagiri, it is necessary to summarise briefly the results of the Brahmagiri 
excavations, a work which was carried out with thoroughness and accuracy. 

Of the three cultures revealed on the occupation-site, phase “ B ” of the 
earliest, the Stone Axe culture, is of importance since 80 out of 102 micro- 
lithic blades and implements have been recovered from its layers. The 


preceding phase “ A ” of the same culture has produced only 9 specimens 
The Megalithic culture which succeeds the Stone Axe culture, has also 
yielded some microliths from its layers, though no more than 9 specimens. 
The chronological position of the microlithic culture of Brahmagiri is thus 
fixed. It would appear that the microlithic culture had its beginnings in 
phase A of the polished Stone Axe culture and survived into the Megalithic. 
Its fullest development was obviously during the “ B ” phase. In fact, it 
would be worth while to test by further excavation the question whether 
the microliths found in “ A ” below, and in the megalithic level above, are 
truly in situ , or whether earth-movements or burrowing animals had trans- 
located them. As regards the four specimens found in the upper strata, 
the Andhra culture level, translocation must be considered virtually certain. 

From a study of the artifacts, another important point emerges, namely 
that jasper is the most frequent among the raw materials used in the making 
of the microlithic implements. Obviously it was highly prized, perhaps 
because of its colour and attractiveness. Furthermore, excellent supplies 
could be procured only 10 to 15 miles away, namely 3 miles north of 
Sandur town among the Sandur hills, and also from the beautiful cliffs 
of Ubbalgondi, a village 7 miles east south-east of Sandur (Foote, 1895, 

Along with jasper, rock-crystal, camclian, agate, flint and common opal 
occur as raw materials. These and other raw materials such as chert are 
available in the district, a fact which renders intentional selection of the 
coloured jasper highly probable. 

The site of Kibbanahalli, which is remarkable for its Palaeolithic artifacts 
(see p 16) has also produced a microlithic industry. From an examination 
of the waste material, it appears to be a factory site too. The site was 
discovered by Mr. and Mrs. Allchin, of London University, who visited that 
area in 1951 in the hope of finding palaeolithic artifacts. There are a 
number of small spurs extending towards the cultivated lands of the village 
of Kibbanahalli. The implements and the waste material all came from 
these spurs. 



Bangalore (1): JalahaJli. 

The Jalahalli industry must be regarded as truly microlithic, since the 
majority of the pieces is smaller than 2 cm. The maximum length for 
lunates is only 18 mm. and the minimum 9 mm. for those found in situ 
at the quarry hill, and 20 mm. and 9 mm. respectively for the surface site. 
The maximum length of pointed flakes blunted on one side or part of one 
side is 22 mm. in the surface site, and 20 mm. in those found in situ. 
Generally speaking, the size of the implements, as Gordon (1938) suggests, 
largely depends upon the raw material, which is mainly quartz. 

In spite of their small size, the implements are remarkably varied in 
type. From a close examination' !) of the specimens in the British Museum 

(1) Grateful thanks are due to Mr Alexander and other authorities of the British Museum for giving me all 
facilities to study Todd's Collection of microliths from Jalahalli Ttie whole assemblage has been studied 
in detail in order to elucidate the character of the industry. 


it is evident that the proportion of worked specimens to unretouched flakes 
is unusually high. About 75% are retouched flakes. There are six cores 
varying from 1 cm. to 3 cm. in length, made of quartz and rock-crystal. 
The majority of the cores from the Bangalore sites are pointed (or “conical”), 
though cylindrical and conical flat-based cores and chisel-ended cores are 
also present. The pointed and chisel-ended cores are of course nothing 
else but finished specimens derived from conical flat-based originals, whilst 
the rare cylindrical cores represent a different type. The presence of thin, 
long blades shows that fluting, though difficult on quartz, was practised with 
considerable success. It should be noted, however, that larger cores from 
which flakes were taken must have formed part of the Jallahalli assemblage. 
None of these has been included in the collection. Perhaps they were over- 
looked by the collector. 

Points : — The characteristic tool of the series is an asymmetrical point with 
a wholly blunted back. Of the total number of 39, 25 are obliquely blunted 
on the left; 14 on the right. One of these (PI. XIII, fig. 51) is worth noting 
because it is the only specimen made of red jasper and also shows signs of 
use (Todd, 1948, fig. 2, No. 29). One point deserves special mention. 
It is made of rock-crystal, obliquely blunted on the right, and may be com- 
pared with No. 23 of Fig. 16 of Clark (1936). No. 33 (Todd) has steep 
retouch and has been blunted obliquely on the left (PI. XIII, fig. 26). No. 50 
and No. 51 (Todd) are about 3 cm. long, blunted on the right, and form the 
largest blades of the whole assemblage (PI. XIII, figs. 64, 65). It is interest- 
ing to note that there is a round-based point present (PI. XIII, fig. 18) and 
a small curved point of rock-crystal (PI. XIII, fig. 33) which is very like those 
called “Frensham ” points in southern England. Such points have been 
found at Spreakley and Bron-y-De on the surface and at Kettlebury in 
excavation (Rankine 1951). 

There are some rod-like artifacts with a triangular or polygonal cross- 
section. They are about a centimetre long and are probably just spalls 
since they show no secondary working. One of them, however, appears to 
have been retouched and probably used as a drill. 

Nos. 42, 43, 44 (see PI. XIII, figs. 47, 27, 49) and 45 (Todd) probably 
fall into the category of drills. No. 44 is of particular interest since it has 
been finely blunted on both sides. 

Triangles : — There are five triangles and two scalenes (PI. XIII, figs. 56, 58) 
in the collection. Nos. 27, 28 and 38 (Todd) are asymmetrical triangles of 
rock-crystal and show fine retouch (PI. XIII, figs. 52, 53, 54). 

Lunates: — Next, the lunates constitute the most conspicuous group of the 
assemblage. Todd notes that they form 38.7% and 32.4% of the artifacts 
from the quarry hill and the surface respectively. Practically none of the 
lunates shows any signs of a bulb of percussion. 

Lunates or crescents may be divided into two varieties. In the one 
the chord is blunted, in the other, the arc. The majority of the lunates 
belong to the latter category (see PI. XIVB). No. 25 (Todd) has the chord 
edge trimmed, in addition to the blunt edge of the arc (PI. XIVB, fig. 4). 
Many of the lunates are asymmetrical, and approach the blunted-back point. 

Petits tranchets : — No fewer than nine petits tranchets have been found at 
Jalahalli and constitute a characteristic type of this industry. The “petit 


tranchet ” (English equivalent ‘ transverse arrowhead ’ (Clark, 1932, p.xxi)) 
consists of a section of a flake with steeply blunted sides. Jalahalli, No. 7 
(Todd) deserves to be specially noted as a good example (PI. XIII, fig. 2). 
On the whole, the transverse arrowhead occurs very rarely on Indian sites. 
Only one other specimen so far has been published, namely by Bruce Foote 
(1916, PI. 14, No. 43). It was found at the teri site of Sawyerpuram in 
the Tinneveily District. There are, however, a few more specimens in the 
recent collection made by Mr. Lai and Mr. V. D. Krishnaswami from various 
sites in the Tinneveily District. 

Burins: — The hall-mark of the burin, a pointed flake, or less commonly a 
core tool, is the burin facet which is produced by a blow struck at the 
working point along the length of the flint, the blow truncating the edge 
of the blade or flake (Burkitt, 1920, pp. 306-310; Clark, 1932 and 1936; 
Noone, 1934; and Garrod and Bate, 1937, Pis. IX, XVI, XVII, XVIII, XX, 
and XXIII). Though the burin is one of the most characteristic types of the 
Upper Palaeolithic of Europe, it is sometimes abundant in the Mesolithic. 
At Star Carr, Yorkshire, for instance, it is the most frequent tool-type 
that can be recognised. 

The Jalahalli assemblage comprises five burins, three of quartz and two 
of rock-crystal. The rock crystal specimens are very small and measure 
13 mm. and 10 mm. respectively (PI. XIII, figs. 60, 61). It must be noted 
that burins of quartz are not so obvious as those of flint or agate owing 
to the less regular surfaces of the flake scars. The rock-crystal specimens 
belong to the ‘ J ’ type and ' L ’ type of Noone (1934) respectively, and 
the quartz burins to the ‘ A ’ type (Bec-de-flute). Two of them were figured 
by Todd (1948, fig. 2, Nos. 52, 53; see PI. XIII, figs 63, 59 in the present 

Scrapers : — Many types of scrapers are present among the Jalahalli imple- 
ments. There are ten end scrapers (including Todd’s Nos. 54 and 58; see 
PI. XIVA, fig. 13) and six side scrapers. Three of the side scrapers (PI. 
XIV A, figs. 10, 11, 16) are longer than broad, finely retouched on one side, 
which is slightly curved. The remaining three are simple thumb-nail 
scrapers (PI. XIVA, figs. 2, 4, 7). Only two nose scrapers are present in 
the collection (PI. XIVA, figs. 15, 17). 

Flakes : — Of the numerous flakes, hardly any show marks of utilisation. The 
largest flake is 5.25 cm. long. It is thick and heavy and has a prominent 
bulb of percussion. It is made of vein quartz and shows no further signs 
of retouch or use. 

Bangalore Sites: (2) and (3) 

A few words should be added about the implements from the other 
two sites near Bangalore. The artifacts from the Hindustan Aircraft factory 
site (site 2) are not representative since the number of implements is 
only 17. Though observations on such a small assemblage are liable to be 
misleading, the group appears to have some resemblance to the Jalahalli 
series. In the first place points are present in both in large numbers 
(some blunted on the right, others on the left). Of the implements from 
the Aircraft factory site, 3 are blunted on the left, 2 on the right. The 
presence of a drill in the series reminds one of Jalahalli once again. Lastly, 
the 12 lunates from this collection (see PI. XV, figs. 20 - 30), over 75% 


of the number of the implements (not counting the cores), bear a close re- 
semblance to those of Jalahalli. There is, so far as evidence goes at the 
present, no objection to the assumption that the same culture is found at die 
Hindustan Aircraft factory site as at Jalahalli. 

The last site (site 3) lies near the Bangalore - Sarjapur road. It appears 
to be a good manufacturing site. In addition to the presence of a large 
number of primary flakes, there are a few cores, parts ot cores and core 
rejuvenation flakes. 

A finely retouched tanged arrowhead (PI. XV, fig. 14) is conspicuous 
among the implements from this site. It is a good specimen, made of chert. 
PI. XV, fig. 1 shows a large, thick point, probably an arrowhead. There 
are blunted-back points, of which PI. XV, fig. 5 is a very good example, 
which has been blunted obliquely on the left all along the edge. The bulb 
of percussion has been removed carefully. PI. XV, fig. 8 belongs to the 
same category. It has very good steep retouch, although the material used 
is quartz. Another one is retouched on both sides at the point and may have 
been a drill (PI. XV, fig. 3). Three others have been blunted obliquely 
on the right, along the whole edge. 

There are two blunted-back “ knife ” blades in the collection. One 
of them is a large blade-flake with the bulb of percussion still present. The 
other one is a fragment only of a blade, of which one side is retouched 
steeply and the other trimmed lightly. 

Of the 15 scrapers, 6 specimens are end scrapers. Some good examples 
of end scrapers or blades are included among the remainder Three steep- 
edged scrapers, which may be small editions of the keeled scraper deserve 
special mention (PI. XVI. tigs. 8. 9. 10' The icm are side scrapers or flakes 
with steep retouch along a curved edge. Some arc longer than broad, but 
others approach the thumb-nail type. These side-scrapers are characteristic 
of the Bangalore industries. 

Only three crescents (PI. XV. figs. 17. 18. 19) and an irregular triangle 
(PI. XV, fig. 2) are present and their number is very small compared with 
the total number of implements from the site. 

This industry cannot easily be distinguished from those of the other 
Bangalore sites and it would be unwise to u. tempt to separate them on typo- 
logical grounds. Moreover, the raw material and the eiat physique are the 
same. They may therefore be regarded as a unit, at least for the time being. 

(4) Brahtnagiri 

No cores were found in situ at Brahtnagiri. Dr. Kiishna's excavation 
produced none, neither did Dr. Wheeler’s in 1947 But seven cores have 
been found on the surface. Of these two are of rock-crystal, two of jasper, 
one flat-based (fluted) core of quartz, one small lump of chalcedony and one 
conical core of banded agate. That Brahmagiri was a chipping floor is 
proved also by the presence of a very large number of primary blades and 
flakes and also of core trimmings and core parts struck off in the course 
of rejuvenation of cores, and found on the surface. 

“ Brahmagiri pre-I The Brahmagiri microlithic industry consists of two 
groups of implements. To the first group belong a few flakes of jasper, 
chert and flint, found on the surface, and not known yet from the excavation. 


Some of them have secondary working and show signs of use (PI. XVTIA, 
figs. 6 and 7). PI. XVIIA, fig. 3 appears to be a scraper. PL XVIIA, fig. 4, 
probably of black chert, is a hollow scraper. PI. XVHA, fig. 1 of flint- 
like chert has marginal tri mmin g. One chalcedony flake may have been an 
awl. It is difficult to fit such flakes into the industry which occurs in the 
excavation at Brahmagiri, because they have certain resemblances to the 
microlithic phase I of Sanganakallu, near Bellary. Most of them are fairly 
large primary flakes with Sttle retouch and there are very few implements 
(Subba Rao, 1948, PI. XIV). It appears, therefore, that this grout) of 
surface finds at Brahmagiri may have preceded the IA phase of the polished 
Stone Axe culture of the Brahmagiri excavation. Furthermore, the evidence 
for flaking on the spot, described in the preceding paragraph, would have 
to be assigned to this phase of the microlithic period of Brahmagiri which, 
for the sake of convenience, may be called “ pre-I.” 

“ Brahmagiri IA, IB.” The second group is characterised by slender, long 
blades which are totally different from the first. All the microliths from 
the excavation belong to this group. Seven types or artifacts of this phase 
have been distinguished (Wheeler, 1948). These various subdivisions can, 
however, not always be justified on typological grounds, because there are 
few retouched implements in the whole assemblage. There are sixty-four 
primary parallel-sided blades. They are not blunted, though use-marks 
occur frequently (PI. XVIIB, figs. 2, 4), and were presumably used as knife- 
blades. Seven specimens of parallel-sided blades have been assigned to a 
separate category, because they are slightly curved at one end. This curv- 
ature is due, however, to the original shape of the core from which the 
blades were struck. 

A group of retouched implements from Brahmagiri is that of the blunted- 
back blades. They are not numerous, six specimens only coming from the 
excavation. One very good specimen was found on the surface (PL XVIIB, 
fig. 6). Three more have the working edge slightly serrated. 

A few of the blades are converted into points by a very gradual oblique 
blunting. They resemble Gravettian points from the Upper Palaeolithic of 
Europe, though this does not imply any real affinity of die cultures. One 
good example comes from the surface (PI. XVIIB, fig. 7). Others have 
their edges wholly blunted. 

No crescents have been found at Brahmagiri, though there is one speci- 
men which may be regarded as a crescentic point or blade. It is wise, 
therefore, not to state categorically that crescents do not occur, but it is 
certain that even if there are any, they must be exceedingly rare. 

Two specimens have been classified as ‘gravers’ ( \Vheeler, 1948, 
Fig. 34, Nos 31, 32) but close examination shows that their shaoe is due 
to chance. There is no clean break in the one which would constitute the 
burin facet. Nor does the facet show any bruise or mark of use. The 
other is a beaked blade such as occur occasionally owing to irregularities 
in the raw material. In fact, burins are entirely absent from the Brahmagiri 
microlithic collections. 

This is a characteristic feature of Br ahm agiri. If one accepts the 
industry as an ordinary microlithic one, one would expect to find burins 
(as well as crescents and triangles). This is illustrated by the proportions 
of burins to microliths in some of the European sites, which are as follows: — 


(1) Abinger Common 1:49 

(2) Farnnam 1:26.5 

(3) Selmeston 1 :40 

But at the Maglemosian site of Star Carr, in Yorkshire, Clark found 
a quantity of burins which considerably outnumbered microliths. The 
nature of the site, which is a fishing settlement with numerous antler har- 
poons made with the aid of the burins, possibly explains this preponderance. 
The absence of the burin at Brahmagiri, therefore, may be interpreted as 
meaning that bone or wood was not worked there to any large extent. 

Nos. 25 and 26 (Wheeler, 1948, fig. 34) were regarded as triangles 
in the excavation report, but they appear to be pointed blades without any 
retouch. No. 26 may have been used. Finally, the class of chisel-ended 
blades (type VI), with Nos. 33 and 34 (Wheeler, 1948, fig. 34) as examples, 
has not much to recommend itself. They are mere primary blades from 
which a short terminal Sake has been removed by accident. 

The vast majority of artifacts therefore consists of simple blades with 
little and sometimes no secondary working. Burins, triangles and crescents 
are absent. Chi the whole, Brahmagiri suffers from a dearth of specialised 
implements and therefore must be regarded as a very simple, non-geometric, 
industry based on small parallel-sided blades. Simplicity, however, does 
not mean that the industry is primitive, it may well be the result of secondary 

(5) Kibbanahalli. 

The Kibbanahalli microlithic industry consists almost entirely of quartz 
cores which are mostly blade cores (PI. XVIII, figs. 1, 2), though flake 
cores are present. Some of the large flakes have been slightly retouched 
and used as side scrapers (PI. XVIII, fig. 8). Other flakes which are small 
have also been used, a few possibly as transverse arrowheads (PI. XVIII, 
fig. 3). 

There is an interesting series of blades, some of which have been used 
as such; a few have been backed and three have been transversely broken 
with retouched ends. Some of the blades are tanged also (PI. XVIII, fig. 5). 

In addition to the side scrapers already mentioned, there are a few 
specimens of hollow (PI. XVIII, fig. 7), double hollow and end scrapers. 

But the most finished artifacts of the whole assemblage are the lunates 
(PI. XVIII, figs. 12, 14, 15, 16). Of the five lunates, none has the chord 
trimmed. A sixth specimen may be regarded as a trapeze (PI. XVIII, fig. 13). 
There is only one poor specimen of an obliquely retouched point. Of the 
128 specimens found at this site, 44 are unused and unworked flakes and 
fragments and 84 are used and worked, including finished implements. 


It is pertinent to note certain salient differences between the Brahmagiri 
and the Jalahalli industries respectively. Even a superficial inspection makes 
it evident that both implement types and technique of producing them are 
strikingly dissimilar. Crescents, burins, triangles and transverse arrowheads. 


all with steep retouch, characterise the Jalahalli industry. Such types are so 
far totally absent from the Brahmagiri series where, moreover, die retouch 
is rarely steep, but usually nibbled. On the other hand, Jalahalli has not 
produced rhe blunted-back and serrated blades, which are so common in 
Brahmagiri IA and IB. In addition, large numbers of parallel-sided un- 
retouched blades 2 inches long have been found. These are reminiscent 
of the long, slender blades of Harappa and Mohenjo-daro. Brahmagiri, like 
Mohenjo-daro, represents a semi-urban culture in which arrowheads, 
crescents and other parts of the hunting equipment are absent, whilst the 
long knife-blade, presumably a good and cheap kitchen-tool, remains as 
almost the only piece of equipment made of unpolished stone. These 
microlithic blades were in all likelihood supplemented by copper tools. A 
copper chisel was actually found in situ in the microlithic layer of the IB 
culture, in Dr. Wheeler’s excavation. Previously, Dr. Krishna (1942a) 
had found a tool made of copper in the level of the Stone Axe culture. 
A few other metal objects are known in addition to these. 

The Jalahalli microlithic industry, however, with its preponderence of 
crescents, points and arrowbarbs is ideally suited to a hunting economy and 


Certain differences can be noted between the Jalahalli and Kibbanahalli 
industries. The presence of microburins, which may be regarded as a waste 
product, is not significant. The absence of the burin in the Kibbanahalli 
industry should be regarded as a differentiating feature. Furthermore, the 
Kibbanahalli industry is characterised by the absence of asymmetrical backed 

Tentatively, therefore, the Kibbanahalli microlithic industry is classified 
as a third group, apart from the Jalahalli and Brahmagiri I industries. 
Whether this new group can be separated from Brahmagiri Pre-I cannot be 
decided until more collections have been made at Brahmagiri. 


In order to arrive at a clearer understanding of the Mysore microlithic 
industries than is possible with the aid of the material found within the 
State boundaries, a number of other sites have been studied. 

Sawyerpuram. The “Teri” ( " sites, including Sawyerpuram (Tinnevelly 
district) are among the most interesting factory sites of the microlithic group. 
Foote (1916) collected 71 chert and quartz flakes and cores from Sawyer- 
puram. Reference has already been made to a transverse arrowhead from 
this collection. Aiyappan (1945b) visited the site in 1942 and obtained 
a further 85 implements, apart from a large number of waste flakes of various 
sizes. In 1949, more artifacts were collected during Professor Zeuner’s 

(1) “ Teri,” a Tamil word, means sand dune The site is covered with fine wind-blown red sand 


expedition. Another series from this and other neighbouring sites was 
collected recently by Mr. B. B. Lai and Mr. V. D. KrishnaswaIni. (l, 

There is some general similarity in the type-tools of Sawyerpuram and 
Jalahalli. Blades, burins, transverse arrowheads, and crescents are found 
in both. In both, some of the crescents have trimming on the chord as well 
as retouch on the arc. In spite of these common features, the two industries 
need not be identical and contemporary. But it appears safe to say that 
they are more closely related to each other than is either to Brahmagiri. 

Bandarawela. The large series of implements from Bandarawela in Ceylon 
may be grouped with the Jalahalli industry in the widest sense. This at any 
rate was Todd’s impression when he reported on the latter locality (1948). 
The present writer has been able to study the Bandarawela material in the 
collections of the Pitt-Rivers Museum, Oxford, and the British Museum, 
London, as well as the Jalahalli material. The resemblance noticed by Todd 
is caused largely by the absence of long blades in both industries. Beyond 
this, however, some differences are noticeable. The Bandarawela series, for 
instance, contains no screw-driver type of burin. On the other hand, Acre 
are composite tools in the Bandarawela industry (N. A. Noone and H. V. 
V. Noone, 1940), some specimens bearing more than one working edge. 
Its range of tools is far greater and types are more varied (60 types and 
varieties are distinguished). 

On the grounds of raw material Jalahalli and Bandarawela are related. 
In both, quartz is used instead of the usual chert or chalcedony. At Band- 
arawela, the raw material is almost exclusively rock-crystal, whilst at Jalahalli, 
common quartz predominates, though rock-crystal is not absent. In brief, 
these two industries both belong to the “ hunting ” microlithic, but are other- 
wise typologically distinct. Resemblances are somewhat stressed by the use 
of similar raw material. 

Langhnaj. Excavations at this site in Gujarat have revealed an industry 
connected with the “ hunting ” microlithic series. Finished implements are, 
however, rare on this site. There are two crescents (Nos. LP.1725, 
LP 1724), one crescentic point (No. LP.1726), and a fine point (LP.1723) 
blunted obliquely along the whole right edge, in the collection of the Institute 
of Archaeology, London. 

Bellary. Many industries distributed in the Bellary District, which lies just 
north of the Mysore border, have close parallels with Brahmagiri. Like 
Brahmagiri IA and IB, each one of the sites to be mentioned below has 
yielded a microlithic industry associated with neolithic celts, adzes and 
chisels. A large percentage of the artifacts consists of unretouched parallel- 
sided blades. Most of the finds are unfortunately unstratified but this 
much is clear — that the Brahmagiri I industry was widely distributed north 
of Mysore. 

The factory site of Kupgal is the most conspicuous. Along with celts 
and blades (Nos. 827-1 to 827-34 <::) ), cores (827-56 to 827-83) — some with 
as many as 12 blades struck off — were found. Such is also the case with 

(1) A detailed study of this new material has been made in Professor Zruner’s Department (Zcuncr and 
Alkhm (1956) The statements made here apply to the previously published material only 

(2) AH the numbeis refer to Foote (1914) 


the North Hill and Fort Hill of Bellary. Kotekalludrug, 8 miles east-north- 
east of Adoni has produced similar evidence (neolithic celt, 1441; blades, 
1440-1 to 1440-16; and 4 cores). 

The Cinder Camp of Lingadahalli, 30 miles north-east of Bellary con- 
tains a series of celts, adzes and hammerstones (Nos. 1464 to 1514) and 
microliths, mostly blades (1515-1 to 1515-86). Twenty-two miles north- 
west of Bellary is the site of Gadiganur which has yielded evidence of 
cultures ranging from the Palaeolithic to the microlithic. Fifty long blades, 
a few having their backs blunted, 14 cores, along with a large number of 
neolithic implements were found by Foote at this place. Another important 
site with microliths and neolithic implements is Kurikuppa, 17 miles west of 
Bellary (celts, etc., 1214-49; microliths, 1258-1 to 41; cores, 1258-48 to 
60). Others showing similar evidence are Halekote, a fine site on die 
Tungabhadra, 31 miles north of Bellary, Ramadurgahill, 20 miles north-east 
of Bellary and Malyam, in the Rayadrug Taluk. 

Much more important than all these is Sanganakallu, near Bellary itself, 
where the evidence comes from stratified deposits. Dr. Subba Rao (1948) 
found a large number of parallel-sided blades in his excavations. This 
industry (Sanganakallu phase II) corresponds to the microlithic industry of 
Brahmagiri IA - IB. 

Anantapur. Lattavaramuhill site (ceks, etc., 2071-82; blades, 2083-12 to 
38; cores, 2083-55 to 88), Jambuldinnehill site, 10 miles north of Anantajtur 
(celts, etc., 2095-2099; flake-blades, 2100 to 26), Kalamedevurhill site (celts, 
etc., 2146-2158, flakes and blades and cores, 2162, 63, 1935-2011. flakes 
and blades, 2040-2 to 2040-8; cores, 2040 to 32) all belong to the Anantapur 
District. Ihev all appear to belong to the Brahmagiri IA-IB group of 
microlithic industries. 

Kumool. The Kutnool District contains a number of sites all yielding celts 
in association with microlithic blades. Among these are Bastipad, Vcladurti 
and Mulagundamu fort hill. The most important site of this kind in the 
Kumool District is Patpad. Here, Foote found a large number of micro- 
lithic blades (2376-95 and 2441-2474) and cores (2475-2604) with celts and 
megalithic pottery of forms identical with those of Brahmagiri. Whilst the 
celts and the microliths may belong together, the megalithic pottery may 
well be later, as it is in Brahmagiri. 

Hyderabad. In Hyderabad, the Lingsugurtaluk of the Raichur District has 
a number of sites, with industries almost identical with those of the micro- 
lithic culture of Brahmagiri. Bellamur Rayan Gudda (celts, etc., 2620-23; 
Hades and cores, 2628-2633), Kotegallu (celts, etc., 2634-38- blades 2639-1 
to 42), Wuttugalluhill (1) (celts, 2640-49; blades and cores, 2665-68), Anan- 
dagal (celts and chisels, etc., 2710-2719; blades, 2724-34) and Rawalkonda 
(celts, etc., 2786-2871a; blades, 2789-2848 and cores, 2849 to 2870) are the 
important sites. Two more sites in the Hyderabad State have close links 
with Brahmagiri, and deserve special mention: Kallur with its copp>er swords 
and microliths, and Maski (2) with its Asokan inscriptions and neolithic finds 
together with microlithic blades and implements. These testify to an 

(1) No 2663 U a remarkably long blade of 3 3/16 inches resembling the Indus valley long blade* (Foote, 1916). 

(2) Annual Report of the Archaeological Department of Nizam’s Dominions (1938) 


identical culture or cultures. Systematic excavation of Maski will reveal 
these cultural links more clearly. 

All these sites conform to the Brahmagiri evidence, as judged by the 
surface finds. More work on them is urgently needed which may show the 
precise relationships between the cultural levels. 

Cutch. A large number of parallel-sided blades, points and retouched 
blades with marks of use like those from Brahmagiri have come from Lodai 
on the island of Cutch and are now in the Sturge Collection at the British 

Nasik. Recently some more interesting evidence has come to light from 
the excavations carried out by Dr. Sankalia at Nasik (1) . The microliths of 
Nasik, too, bear a similarity to those of Brahmagiri. The evidence of 
pottery and copper axes from Jorwe, not far from Nasik, fits in with the 
general picture. 

It is thus clear that the microlithic industry of Brahmagiri IA - IB had 
a wide distribution in South India (see PI. IV A). 

What has been said so far is based on the published evidence (except 
for Sanganakallu I which will be discussed presently). There are other sites 
from which collections have been made but not published which show that, 
apart from the Brahmagiri I (which appears to dominate north of the Mysore 
border judging from the published sites alone), there are industries which 
do not agree with those described. 

Sanganakallu. Phase I is comparable to Brahmagiri Pre-I. This precedes 
Brahmagiri IA and IB on the one hand, and on the other, is distinct from 

Jubbalpur. This does not conform to any of the three. It appears to be 
a mixture and requires to be investigated. 

To sum up, Mysore has yielded four types of microlithic industry: 

(1) Jalahalli hunting type. 

(2) Brahmagiri IA and IB urban type. 

(3) Brahmagiri Pre-I type. 

(4) Kibbanahalli hunting type. 


The age of the Mysore microliths now remains to be discussed. 

At the very outset we are confronted with the fact that in many places 
in India the microliths do not seem to represent a “ Mesolithic ” culture, 
as they would if found in Europe. Some are clearly associated with the 
neolithic complex. Others may have been proto-neolithic, as De Terra 
and Paterson (1939, p.320) suggest. Microliths from Pachmarhi, Hos- 
tangabad, Singhanpur and Kabrapahar in Central India are late survivals 
and are not likely to antedate 500 B.C. at the earliest (Gordon, 1938, p.23). 
Cammiade (1924, p.101) points out instances where microliths have been 
found in association with proto-historic urn-burials. 

(1) Unpublished. 


Todd, on the other hand, believed that the microliths from Khandivli, 
judging by typology and patination, were much older than those of the Central 
provinces. These finds are outside the subject of this paper, but it is impor- 
tant to realise that some true “Mesolithic” phase may still be established 
in India. On the Sawyerpuram microliths, Foote’s (1916) opinion was that 
they lay embedded in the fossil-sand dunes (tens) and were stained red owing 
to their long contact with the ferruginous soil. Aiyappan, however (1945b, 

р. 146) who recendy re-investigated die site, is inclined to believe that the 
implements were made only at the time when the basal level of the sand 
dimes (teris) was being laid down and not afterwards. He believes that 
the site was deserted by the makers of the microliths before the advent of 
die neolithic technique, since no neolithic tools have been found so far on 
the site or in its neighbourhood. The Sabarmati sites like Lanahnai, 
Akhaj, Valasna and Hirpura, discovered and excavated in Gujarat by Sankalia 
(Sankalia and Karve, 1949) have produced implements in association with 
mineralised human skeletons. In fact complete skeletons of ‘microlithic 
man ’ have been found. Here, too, the main series of microliths belongs to 
the pre-pottery layers. 

Thus die microliths of India cover a lon^ range of time. Their begin- 
nings, indeed, may go back to a period which may one day prove to be 
placeable in the Mesolithic, but they survive well into the historic period, 
on the evidence which is available to us at the moment. 

In considering the age of the Mysore microliths in terms of this back- 
ground, the Jalahalli and the Brahmagiri industries must be taken separately. 

The chronological evidence for Brahmagiri is fairly clear. The micro- 
lithic industry is associated, in the main, with polished stone celts of the IB 
and IA phases of the Axe Culture. They disappear above the layers of 
the Megalithic Culture, which may be safely dated to the Maurya period 

с. 300 B.C.). The polished Stone Axe Culture at Brahmagiri has been 
dated to c.1000 B.C. (Wheeler, 1948) on carefully considered evidence. 
The microliths of Brahmagiri, therefore, range in point of time between 
c.1000 B.C. and c.300 B.C. This provisional dating based on present 
stratigraphical evidence will have to be corroborated by further study of the 
‘A’ phase of the Stone Axe Culture. 

In view of the slender evidence, the dating of the Jalahalli microlithic 
industry presents a more difficult problem. Differences in the implement 
forms from those of Brahmagiri possibly suggest a very different date. 

The implements were found in a layer, above the granite and below 
a black soil, composed of pellety laterite and rubble of the decomposed 
granite itself (Todd, 1948, p.27). The chronology of the industry at this 
site, in the final analysis, can only be settled, if at all, by a careful examina- 
tion of this soil layer. The occurrence of the tools in the pellety laterite 
layer appears to indicate an early age for this industry, since it lies directly 
above the bedrock. The black soil, one foot above this layer, produced 
pottery. Whether this interval of one foot between the implementiferous 
layer and the pottery layer suggests a break in the continuity of these two 
cultures requires further investigation. 

Todd believed that, owing to the absence of heavier types of tools, 
the Jalahalli industry belonged to the beginning of the latest microlithic phase, 


but tbe occurrence of a large quartz Bake suggests the possibility that heavier 
tools are included in the equipment, and that their presence had escaped the 
notice of collectors. 

Rankine (1951) suggests the following characteristics as defining the 
Mesolithic (see PL XII): 

(1) Slender blades of varying lengths and widths with edges roughly 

(2) The making of certain microliths by the notch method. This 
is an ingenious device whereby the bulbar extremities could be 
removed and a retouched point developed in the strongest section 
of the blade. 

(3) Use of a steep retouch known as blunting and a lighter retouch 
known as trimming. 

(4) Rejuvenation of cores and resharpenin g of tranche! axes. 

Of these criteria, the last does not apply to Jalahalli for want of re- 
juvenated cores and complete absence of tranchet axes. A petit-tranchet 
is present, however, with an excellent tranchet blow across die tip. Points 

(1) and (3) apply in part. 

Furthermore, the presence of burins has often been regarded as indi- 
cating a Mesolithic (or even Upper Palaeolithic) culture. Blade-and-burin 
industries have been reported from three regions in India: 

(1) In and around Bombay (Todd, 1939). 

(2) Madras Presidency (Cammiade and Burkitt, 1930). 

(3) Jalahalli, Mysore State. 

Burkitt’s stage IV from the Nandikanama Pass is possibly another site 
of this kind and similar to Jalahalli. On this evidence it would be possible 
to maintain that a Mesolithic industry existed in India, were there not 
indications that the characteristics commonly given for the Mesolithic con- 
tinued into later periods in certain regions. 

Thus, Todd (1948) believed that the burin indicated an early date for 
the Jalahalli series. But instances are not wanting for the late persistence 
of the burin. For example, a series of burins including angle, bec-de-flute 
and busque, occurs in the neolithic Dhobaian industry of Palestine (Waechter 
and Seton-Williams, 1938). 

Outside India, the Wilton A and B industries from South and East 
Africa are strikingly like Jalahalli. Both at Jalahalli and in the Wilton sites 
are found: 

(1) a large number of crescents 

(2) a large number of thumb-nail scrapers 

(3) points and burins. 

One difference has, however, to be noted. No pottery is found asso- 
ciated with the implements at Jalahalli. The Magosian and Wilton Cultures 
are both associated with pottery and ostrich egg-shell beads. Wilton A and B 
belong to the Makalian wet phase and B continues at least into the Nakuran 
phase which has been approximately dated to 1000-850 B.C. bv Leakey. Some 
African authorities even hold that the Wilton continued right to die arrival 
of the Europeans. 


The age of the Jalahalli artifacts, therefore, cannot be determined 
conclusively on typological grounds. Geological evidence, however, seems to 
assign a greater antiquity to Jalahalli than Brahmagiri, for the implementi- 
ferous horizon of Jalahalli is covered by a sterile level and by one with 
“ Iron Age ” pottery. It would appear, therefore, that four microlithic 
phases can be tentatively distinguished in Mysore State: 

(1) Jalahalli (2) Br ahma giri Pre-I (3) Brahmagiri I A - B and (4) Kib- 
banahalli. The argument for the distinction of (2) and (3) has been put 
forward on p. 35. Typologically the four phases are different, too, though 
the material from (2) is scanty. Industry (3) is plainly of the “ Urban ” type 
and (1) and (4) belong to the “ Hunting ” group. One must not, however, 
regard this ecological difference as indicating evolution. ‘ Hunting ’ and 
‘ urban ’ microlithic complexes have existed in India side by side from the 
time of the Indus valley civilisation to the Iron Age. 


Chapter IV 


A. Introduction 

The neolithic culture of India presents a complicated problem. Many 
so-called “neolithic” artifacts have been collected, and there is hardly any 
museum in India which does not possess them. But the field evidence to 
define them clearly as neolithic is very inadequate. The term “neolithic” 
is nearly always used in India to designate the presence of polished axes. 
They are very numerous, and there is a tendency to regard the neolithic 
Indians principally as tool-makers (Chakravarti, 1944) forgetting that this 
may have been only one aspect of a complex culture. The neolithic Indians 
need not have been inferior to the neolithic Egyptians, who, in addition to 
tool-making, knew agriculture, domestication of animals, manufacture of 
pottery and textile industry. 

The evidence for activities other than tool-making may well have been 
overlooked and it is not beyond the bounds of possibility that careful exca- 
vation will uncover them in the dry earth of caves, rock-shelters and other 
sites. Indeed, pottery has been found in abundance at Brahmagiri, Sangana- 
kallu and elsewhere. Neither is it impossible that a proper identification of 
excavated animal bones will produce proof that some of the species were 
domesticated (Worman, 1949). Vavilov’s theories (1926; see also Watkins, 
1933) of the history of cultivated plants, which are based on regularities in 
the distribution of crops and on die reconstruction of centres of diversity, 
suggest that India lay not very far from one of these, namely Afghanistan. 
Hence, cultivation of plants, particularly wheat ( Triticum dicoccum and 
Tnticum vulgare), may have been known in India since early times, possibly 
with the advent of the “ neolithic ” artifacts. 

Neolithic celts are widely distributed in India. The first was discovered 
in 1861 by H. P. Le Mesurier in the valley of the East Tons river in United 
Provinces. Theobald (1862) and Cockbum (1879) explored a large number 
of sites which yielded polished stone celts in the Banda District. Ranchi, 
Santal Parganas and the Singbhum Districts of Bihar contain many sites. 
In 1872 Fraser discovered a celt-factory on Kupgal hill, Bellary District. 
Subsequently, Foote explored a very large number of sites in Salem, Arcot, 
Cuddapah, Kurnool and other parts of South India. Hyderabad, too, is rich in 
sites of the polished Stone Axe Culture. Mention must be made of Kotegal 
and Anandgal in the Raichur District, which appear to be celt-making 
centres. 01 Last but not least, there are the various districts of Assam, 
which are studded with stone-axe sites. 

Mysore also abounds in sites of the neolithic group. Foote (1916) 
picked up stray specimens at Birmangala and Srinivaspur in the Kolar 
District and at French Rocks, T. Naraspur and Lakshampur in the Mysore 
District. Polished stone celts have been found in situ and have been exca- 

(1) Foote’s collection 7710 to 2719 (1914) 

Col. D H, Gordon collected some fine specimens, which are now in the Institute of Archaeology, London. 


vated from two sites in northern Mysore, namely Chandravalli and Brahmagiri 
(see PL V). 

If these sites are plotted on the map, they appear to be grouped mainly 
in three dusters, namely: — 

(1) Eastern India 

(2) Central India 

(3) Southern India. 

This distribution of the sites is at the base of a theory put forward by 
Mr. Eugene Worman (1949). Polished celts being confined mainly to the 
eastern half of India, he postulates that the culture to which they belong 
was derived from the Far East He is almost certainly right as regards 
one particular type of axe, the so-called shouldered celt. It is small, flat, 
rectangular and usually shouldered for hafting purposes. Its main area of 
distribution lies in Burma, Indo-China and other parts of south-east Asia. 
In India this type of celt is clearly intrusive and is limited in its distribution. 
V. Ball (1875) found this type of celt in Dalbhum, Chota Nagpur District 
A few come from the Tezpur District in Assam (Coggin Brown, 1917). 
Cammiade (1930) found a stray specimen on the Godavari. Mayurbhanj has 
produced some specimens too. Baidyapur, 14 miles south of Baripada 
yielded two small specimens. Another somewhat larger celt comes from 
Amsikra, about three miles south-west of Baripada town. (1) Worman 
(1949) picked up another in northern Mysore, without, however, specitying 
the exact locality. One of the specimens of the Haimendorf collection of 
stone implements from Hyderabad, in the Department of Environmental 
Archaeology, London University, appears to belong to the same type, al- 
though shoulderless. 

These curious celts are so unlike anything ever made by man of the 
true Stone Age, and on the other hand so much like blades of metal axes 
or adzes, that it is difficult to combine them with the ordinary “ pointed- 
butt ” and similar types to form one “ neolithic ” culture complex. They 
are almost certainly stone copies of metal tools. That this should be so 
is by no means extraordinary, for, in Europe it is common to find in the 
Bronze Age carefully-made copies of metal axes in flint or some other rock. 
These specimens even occasionally exhibit the casting seams of their metal 
originals. It may, therefore, be presumed that the shouldered-celt culture is 
a late intrusion into India from the Far East, and that it is contemporary 
with one of the metal ages. So far as these “ stone axes ” are concerned, 
Worman’s theory applies, although the culture should not be called neolithic. 

Haimendorf (1943, 1945), correlating the megalithic culture of Assam 
with that of the Gadabas and Bondos of Orissa, points out that they all sneak 
±e same form of Austroasiatic tongue and believes that a late neolithic 
civilization with eastern affinities and associated with Munda-speaking peoples 
permeated the older population of the Deccan. This older stratum of the 
“ neolithic ” of India is perhaps represented by the pointed-butt type of 
stone axe. This is the most widely distributed type and forms the majority 
of the finds from more than two hundred sites spread over South India. 
The pointed-butt axe is not, however, uniformly distributed in India. It is 

(1) Out of these three, two arc in the Baripada Museum, information kindly communicated by Mr P. 
Acharya, Mayurbhan) 


crowded, for instance, in the Bellary and Raichur districts, and very rare 
or absent in regions on both the coasts. One is tempted to interpret such 
uneven distribution in terms of racial movements, as indeed Haimendorf 
has done. But it must be noted that the distribution of these axes is clearly 
dependent on the occurrence of a suitable raw material, mainly trap rock. 
As examples, the sites of the Bellary, Raichur and Chitaldrug districts may 
be quoted, which are based on trap dykes. This correlation of axes and 
dykes was already remarked upon by the mapping geologist of the Bellary 
District, Bruce Foote (FI. II). 

The absence of pointed-butt axes in certain areas, therefore may be 
explained in one of two ways. Whilst the raw material was lacking in 
any case, the appertaining culture may either have been absent in those 
areas, or alternatively it may have been present but devoid of the axes. 
The former alternative is somewhat supported by HaimendorFs Munda 
hypothesis but the latter cannot be ruled out. It will have to be tested 
with care by the comparison of the pottery accompanying the pointed-butt 
axes where they occur, with pottery from early sites in the areas without 
these axes. For this reason it is worth while here to review briefly, the 
material culture encountered in a few “ neolithic ” sites. 

At least three sites have revealed some information regarding the 
material culture of the “ neolithic ” people, namely: 

(1) Bursahom (Burzahom) in Kashmir 

(2) Sanganakallu,near Bellary, close to the northern boundary of Mysore 

(3) Brahmagiri in northern Mysore. 

Bursahom. The trial excavations carried out by De Terra in 1935 near 
the village of Bursahom on a table-land with a group of megaliths revealed 
a succession of three cultures. 

III. contains sherds of the Buddhist period of the fourth century. 

II. contains predominantly black polished ware with incised and 
geometrical designs belonging to the Jhangar phase of the Indus 
valley civilization. Apart from hearths and charcoal, fragments 
of polished greenstone occurred also. 

I. revealed grey hand-made pottery in association w'th polished celts. 
The pottery showed textile-and-matt design. The polished celts 
and the pottery were embedded in a yellow loessic soiL 

De Terra (1942) further obtained 30 artifacts of all sorts from the 
surface. These included two celts; one of trap and the other ot amohibol'te. 
This north-western site is of great interest, for it suggests on stratigraphical 
evidence that a polished Stone Axe Culture may antedate the Jhangar phase 
of the Indus valley civilization. 

Sanganakallu. This site, which is about three miles from Bellary, is of 
great interest and has thrown some new light on the problem of the “ neo- 
lithic ” culture in South India. Trial excavations carried out by Dr. Subba 
Rao produced some 30 tools, including celts in various stages of manufacture, 
in association with coarse hand-made pottery. A dull brown ware occurred 
along with a pale-grey ware. The dull brown ware has violet and purple 
paintings (Subba Rao, 1948). The sherds of this painted ware hsv a 
striking resemblance to those from Brahmagiri, to be described presently. 


As at Brahmagiri, they were found associated with coarse, grey hand-made 
ware. Nos. 5 and 6 of Subba Rao (his PI. VU, 1948) have single violet 
lines on a brown background, and are comparable to T16 of Brahmagiri 
(Fig. 18, Wheeler, 1948). Sanganakallu No. 8 has a line and circle orna- 
ment. Nos. 4 and 9 are rim pieces of burnished bowls with red ochre 
paintings; similar specimens also occur at Brahmagiri. No. 7 has a wide 
violet band with cross-hatching and a criss-cross design below. A comb 
pattern occurs on No. 3. The carinated and round-based small bowl has 
also a comb pattern painted all round the shoulder in red ochre. 

These specimens are of considerable interest since they suggest cultural 
affinities between the peoples of Brahmagiri and Sanganakallu. 

Brahmagiri. The last of the three sites is located in northern Mysore, the 
region of the present survey. Careful excavation on this site by R. E. M. 
Wheeler (1948) has revealed some interesting details of the life of the 
“ neolithic ” people. 

The site was in continuous occupation possibly from a microlithic 
period called “ pre-I ” (see p. 34), certainly from the “ neolithic ” onwards to 
the early historic periods. The polished Stone Axe Culture is found in 
deposits up to 9 feet thick and is sub-divided into the phases IB and I 4. It 
is significant that there is a sterile layer in one of the excavations (BR.21) 
between the IB and IA levels of Stone Axe Culture. 

Dwellings : — Little information is available regarding the dwellings of the 
Stone Axe people of Brahmagiri. No structures were found and, owing to the 
limited character of the excavations, no plans of the houses could be ascer- 
tained. The occurrence of post-holes, however, shows that the dwellings 
were probably partly made of timber. Perhaps low walls of rough granite 
blocks were used in addition. The occurrence of a straight line of post- 
holes in one of the cuttings suggests that some of the dwellings were rect- 
angular in plan. 

Disposal of the dead — The Brahmagiri excavations have thrown some light 
on the burial practices of the “ neolithic ” people. Two forms of burials 
were known to them, namely inhumation and urn burial. 

Little information is available regarding inhumation since only two 
examples occurred. One of them was completly uncovered, showing the 
skeletal remains of a child of 8 to 10 years of age. The head of the child 
faced east, and the presence of a spouted pottery vessel near its 
head and two small cups near its thighs suggests that rites of some kind 
had been performed. Whether the spouted vessel was used for pouring 
libation into the mouth or ears of the dead cannot, of course, be decided 
on the basis of a solitary example. 

Urn burials, on the other hand, were frequent. They were apparently 
confined to infants. The bodies were closely packed into coarse, wide- 
mouthed urns and buried. It is interesting to note that this practice of urn 
burial in confined to phase “ B ” of the Stone Axe Culture. All the 14 urns 
uncovered so far belong to this phase. One of them contained a bronze 
pin 3 inches long and also two small pots (T36 a and b), (1) the significance 

(1) These numbers preceded by a letter T, which occur on this and the following pages in connection with 
the pottery from Brahmagirt, refci to the figured specimens in Wheeler (1948) 


of which is not clear. The bronze pin, however, may prove to be very 
important from the chronological angle, for it clearly shows that the Stone 
Axe Culture can be contemporary with Bronze Age Cultures, in spite of the 
fact that at Bursahom, it precedes the Indus Valley civilization, if De Terra 
is correct. Either, therefore, the Stone Axe Culture of the south is later than 
that of the north, or if one assumes them to be contemporary, bronze would 
have been known prior to the Indus valley phase. The first alternative is 
the more likely. 

Pottery: — Before considering in detail the various types of Stone Axe pottery, 
certain characteristic features may be stressed: — 

(1) Throughout the two phases of this culture, namely, A and B, the 
pottery is hand-made, in contra-distinction to the wheel-made pot- 
tery of the later cultures. 

(2) The predominant ware, which occurs in both the phases, is of a 
coarse grey fabric. 

(3) Though most of it is crudely made, same sherds show polish. 

(4) A dominant type throughout is that of a round bottomed vessel 
with plain, slightly everted rim. 

(5) Phase A, in addition to coarse grey potsherds similar to those of 
B, contains painted pottery and incised pottery. These two classes are, 
so far as is known at the present, absent from phase B. 

Painted pottery of phase 1 A — It mus: be noted that the painted sherds 
are too fragmentary to reconstruct pot-shapes. Two types can, however, be 
recognized easily, namely: (i) sherds with red slip, and (ii) sherds with buff 
slip. The difference is not just one of colour. The techniques are different. 
The red slip sherds are salt glazed and burnished. In the case of the buff- 
slipped ones, neither salt-glaze nor burnishing occurs. The painted decora- 
tion, usually executed in ochre is in both cases applied after firing. 

The decorative patterns are: — 

(1) curved lines (T.6). 

(2) plant design (T.3). 

(3) parallel lines (T.2, T.9 and T.12). 

(4) combination of curved and parallel lines (T.l). 

In some cases decorative patterns are formed by incisions, which are 
confined usually to criss-cross and herringbone designs. 

Painted pottery from the surface: — Apart from the painted sherds found 
during the excavations, a number were picked up from the surface, repre- 
senting the three types, with red, grey and buff slips respectively. Six 
pieces are red-slipped and one has a wavy or zig-zag decoration in violet. 
Two are pinkish-buff and decorated with violet parallel lines Another ir'nk sh 
potsherd has a pattern of minute rectangles in brown. Two buff-slipped 
sherds have dark violet bands running in a criss-cross fashion. A perforated 
sherd of coarse dull grey ware was also picked up from the surface. This 
group of finds, therefore, may be regarded as contemporary with Phase 
IA of the excavation. 

Undecorated pottery from phase ‘ A — The undecorated pottery from 
phase A has been classified by Wheeler (1948) according to the rim types 
as follows: — 


i WBtrwnjVu , % »«<«. 

Type I: Rim pieces of dull grey jars with flaring mouth (T.25, T.26). 

Type II: Rim pieces of dull grey vases with flaring mouth (T.27, T.28 

and T. 29). 

Type III: Rim fragment of red ware slightly burnished with a flaring 

mouth (T.30). 

Type IV: Fragment of a rimless bowl of grey ware with light brown 

slip (T.31). 

Type V: Rim fragment of a bowl of grey ware with an external 

groove just below the mouth (T. 32). 

Type VI: Spouted vessel of coarse ware with a thin terracotta red slip 


Type VII: Neck fragment of a bottle-necked vessel of red ware with a 

slightly beaded rim (T.35). 

Comparison with Sanqanakallu : — It has already been pointed out (p.46'1 that 
the painted ware of Brahmagiri IA resembles in many respects that of Sang- 
anakallu near Bellary, which is manifestly another site of the Polished Stone 
Axe Culture. This correlation is corroborated by the undecorated phase A 
pottery of Brahmagiri, which contains types that occur at Sankanakallu also 
(Subba Rao, 1948, Pis. 6 and 9). 

‘ B ’ Phase pottery : — Pottery of the ‘ B ’ phase can be divided into (a) non- 
burial pottery and (b) burial pottety. 

The non-burial pottery and the burial pottery are of coarse grey ware. 
Occasionally in the former class, a black polished ware also occurs (examples, 
T.68, T.63) in the upper layers. The principal types are: — 

(1) vase of dull grey ware with flaring rim and globular body (T.48). 

(2) deep bowls, black and brown (T.60, 62, 61, 65, 67 and 68). 

(3) shallow bowls of dull grey (T.73). 

Burial pottery . — A special coarse pottery was used for burial purposes by 
the Stone Axe people of Brahmagiri. These are mostly urns used for the 
burial of infants. Generally, they are hand-made, of dull mottled grey 
colour, coarse and micaceous in texture, and because of their non-utilitarian 
character are indifferently baked. They have a globular body with a wide 
mouth, flared rim and rounded base. At the mouth the diameter averages 
13 inches. 

Next to the urns come the bowls used for covering the urns. They 
are of two categories, namely: — 

(1) lipped bowls of coarse grey colour (T.44). 

(2) blunted carinated bowls of coarse grey fabric. 

To the third category belongs the funnel-spouted vessel of coarse grey 
ware found beside the skull of the inhumation burial. 

Lastly, the two small cups (T.46a and 46b) of coarse grey ware, which 
were placed near the ends of the two femora of the skeleton, form a sub- 
group by themselves. 



Some Indian prehistorians cherish the opinion that South India never 
witnessed a Copper or a Bronze Age, but came to possess a knowledge of 
copper and copper-smelting only after iron had come into general use. As 
early as 1887 Bruce Foote published the theory that iron had been intro- 
duced among the people living in southern Deccan and iron tools were 
manufactured by them at the same time that they were making and still using 
implements of polished stone. He (1916) concluded that the neolithic 
people were a land-loving people as, if they were sea-faring, they would 
certainly have sailed to die Tenasserim coast through the Bay or Bengal 
and would have become acquainted with the cassiterite of that region. V. 
A. Smith (1905) merely elaborated this idea and held that the Neolithic Age 
was directly succeeded by an Iron Age. As late as 1936, N. G. Mazumdar 
reiterated die same opinion. When Foote and Smith wrote there was hardly 
any evidence contrary to their opinions. Since then, new discoveries have 
been made and in the light of these a chalcolithic phase in South Indian 
prehistory can be confidently postulated and established. The “neolithic 
culture ” of Mysore is in fact contemporary with it. The evidence for raw 
material also leads to this conclusion. The Bellary and northern Mysore 
areas have important copper mines. Captain Newbold discovered many 
ancient copper min es on die Sugalammakonda, near Bellary. Gordon (1952) 
has noted the distribution of the ancient copper mines of South India. 
It is not surprising that where copper ores were readily available, they were 
used by man for his tools. Whether natural copper was used as such, there 
is no indication. 

The first chance-discovery was made in 1938-39 in the vicinity of a hill 
near Kallur, a village in the Raichur District of the Hyderabad State. 
There, while blasting a huge boulder, some labourers found three copper 
swords. The authorities of the Archaeological Department of Hyderabad 
became interested since the swords resembled the Fatehgarh swords of 
copper, (1) 2 3 found previously in the United provinces. Realising the impor- 
tance of Kallur, the Archaeological Department of Hyderabad made a 
survey of the area near the hill where the swords were found. Their survey 
resulted in the discovery of a few early metallurgical sites. A small-scale 
excavation was also carried out at a place called Yammigudda, with impor- 
tant results. A broken copper axe, and a few copper fragments, a lump of 
copper pyrite in association with an old furnace were revealed by the exca- 
vation. With these copper objects occurred microliths. The occurrence 
of the furnace and the copper pyrite testifies that the copper objects were 
locally made. Higher up the hill, a saucer-shaped stone 4' x 4' showed 
traces of copper oxide, copper ore having been crushed on it for a long 
tim e < 2 > 

Fresh evidence has come from a place called Jorwe, about 150 miles 
North-east of Bombay, where 6 copper axes were found in one of the 
painted spouted jars. £3) Dr. Sankalia (1950) found identical pottery, painted 
and spouted, below his layer 5 at Nasik, 128 miles northeast of Bombay, in 

(1) Indian Antiquary (1905) p 232. 

(2) Annual Report of the Archaeological Department, Nizam’s Dominions, 1937-40 p.27. 

(3) Information by courtesy of Dr Sankalia. 


association with blades and cores of chalcedony and chert. That die 
copper tool makers used microliths is thus evident. 

The Jorwe pottery associated with the copper axes occurs below a 
layer at Nasik, which produced the N.B.P. (northern black polished ware) 
which is generally assigned to a period between the third and sixth centuries 
B.C. ll) This N.B.P. is commonly found on most of the northern sites in 
India like Kausambi, Hastinapura, Rajagir and Rajaghat from early layers. 
This datum line might tentatively be used to fix the chronology of the 
painted ware of Jorwe, coupled with its microliths and copper objects. 
A modest date would be seventh to eighth century B.C. for the chalcolithic 
industry, if not earlier. 

We now come to the evidence for a chalcolithic phase in South Indian 
prehistory, and especially Mysore (Lai, 1949). The Mysore evidence is 
particularly valuable because it proves that copper preceded iron. The 
excavations at Brahmagiri produced a copper chisel (from Br. 21) from the 
IB layer of the polished Stone Axe Culture. A thin bronze rod or pin 
was found in association with an urn burial of the same culture. This 
culture, therefore, was characterised by the combination of : — 

(1) polished stone axes. 

(2) microlithic blades. 

(3) copper objects. 

(4) the conspicuous absence of iron. 

The evidence was further corroborated and confirmed by Br. 17. From 
one of its early layers a copper rod was unearthed This cutting, too, did 
not yield any iron object. 1 2 -' 1 

Brahmagiri also throws some light on the disputed question of copper 
alloy (bronze). A chemical analysis of the rod or pin already mentioned 
revealed the presence of 9% of tin. This shows that the preparation of 
bronze alloys was not unknown m the Brahmagiri IB culture. One could, 
therefore, legitimately call this culture a chalcolithic or even a Bronze Age 
one. Nevertheless, me L al tools are rare and, economically, this culture 
was still essentially “ neolithic.” As to IA, no metal has been found, but 
in view of the small size of the excavation, this need not be significant. 
Whether IA represents a “ true neolithic ” culture, therefore, remains to 
be decided. Since both IA and IB are technologically mainly “ neolithic,” 
this term is retained here for the time being. 


It is hardly too much to say that the axe is one of the basic inventions 
of man. It has its forerunner in the cleaver which, functionally, is an axe, 
though it has no artificial handle, the handle being replaced by the arm 
of the user. The concept that “ tools are extensions of the body ” is well 
illustrated by the development of the artificial handle, which first appears 
in the Mesolithic (Thames “ picks ” were probably hafted) and is invariably 

(1) For want of evidence, this dating is accepted here in a preliminary manner 

(2) Ibis evidence ts quite nt with that produced by txcav uions both in and l‘M2 whin the 

wntci and Dr Knshna carried out excavations on the same sue 


present in the neolithic. The deer antler appears to have been the usual 
pattern from which the hafted neolithic axe is derived. 

Antler axes appear to be quite early. For example, the Lyngby axe 
(the “ antler-adze ” of Childe) is considered to be the earliest in Europe by 
Clark (1936) and its date is between 9000-7000 B.C. (Zeuner, 1952). 

Polished axe-like stone tools have been found in large quantities in 
southern India. Usually they are called axes or celts but it is important 
to distinguish several functionally different types (Childe, 1950b). In an 
. axe the cutting edge is parallel to the handle, in an adze it is right angles 
to it. Adzes therefore usually have asymmetrical cross-sections of cutting 
edges. In Mysore the majority of “ celts ” are true axes, but adzes are 
* by no means absent (PI. XIX, fig. 1). 

In addition to celts “ neolithic ” cultures yield chisels which in all 
probability were not hafted. They appear to be rare in Mysore, for none 
has so far been published. The writer succeeded in collecting a broken 
chisel at Brahmagiri. This interesting specimen (PI. XIX, fig. 2) was polished 
all over the preserved portion. The cutting edge is almost half an inch 
long, which is about hall the width o i the boay of the chisel and it is shehtly 
convex. There are use-marks on the edge, so this tool appears to have 
been fractured while being used. Outside Mysore, chisels have been found 
at a number of sites, especially those of the Bellary area. In view of the 
proximity of these to Mysore, the scarcity of chisels in Mysore should not be 
taken as a basis for far-reaching speculation; it may be accidental. 

Other neolithic equipment comprises hammerstones, which may be 
grooved for hafting (Subba Rao, 1948, PI. 23, fig. 20-25), or more frequently 
ungrooved. The latter grade unto “pounders” (Wheeler, 1948, PI. 116) 
which often are spherical. These occur in fair numbers at Brahmagiri, 
whilst grooved hammerstones appear to be rare in Mysore. 

A type of stone-tool called “ fabricators,” made from long flakes but 
- supposed to have been used after the fashion of a baton to pound the sides 
ot rough-ouls when manufacturing stone celts, has been described by Subba 
Rao (1948) from Sanganakallu, but has not yet been found in Mysore. 


Even a cursory examination of a group of celts shows the various 
stages of their manufacture. At Brahmagiri in northern Mysore the writer 
has been able to establish the same process of manufacture wtveh has been 
described from Sanganakallu near Bellary (Subba Rao, 1948). A large 
number of axes has been collected at these sites both from the surface and 
excavations, in all stages of manufacture. Along with these specimens there 
also occur quantities of waste flakes at both places, which clearly show that 
these were factory sites. 

Foote (1916) distinguished four stages of manufacture, namely. — 

(1) flaking (called by him “ chipping ”). 

(2) pecking. 

(3) grinding. 

(4) polishing. 


Convenient nodules of rock were taken and subjected to a process of 
flaking. From both faces or one face as the occasion demanded, coarse 
or primary flakes were removed from the core, to give it the approximate 
shape desired. 

Of these four stages (3) and (4) are difficult to distinguish, since 
specimens that have been resting in the soil will have suffered chemical 
corrosion, whilst specimens embedded in sediments such as hillwash formed 
on slopes, will have suffered abrasion. In fact polish is usually observed 
on very fine grained rocks which lend themselves naturally to the process, 
but celts made from the coarser rocks may never have gone through a 
special polishing stage after the grinding. Subba Rao (1948), too, admits that 
final polishing was applied to “ a few selected specimens ” only. 

The pecking as described by Foote is readily observed on many speci- 
mens, for instance at Brahmagiri (PI. XX, fig. 1). It appears, however, that 
the sequence (1) - (2) - (3) - (4) was often modified by alternating the pecking 
and grinding. This is shown for instance by a celt from Anandagal, 
Hyderabad, collected by Col. D. H. Gordon. This specimen is finished 
as regards its shape, no trace of the flaking of stage (1) being left on its 
surface. It has a regular, sub-circular cross-section. Except for the two 
faces of the cutting edge, the entire surface is covered with batter-marks, 
plain evidence that it was pecked all over. It may therefore be suggested 
that pecking was deliberately applied to surfaces ground smooth, because 
it woud reduce the resistance of the surface to further grinding. The grind- 
ing process would thus have been accelerated considerably. 

It may well be that alternate pecking and grinding was applied many 
times over in the case of well-finished implements. 

The making of a polished celt being a long drawn-out process, much 
care was exercised in the selection of lumps of raw material of suitable 
shape, many being made specially, i.e. struck from the rock as primary 
flakes. Foote (1916, p.18) already pointed out that skilful use was made 
of joint planes in the rock, in order to save labour. This was obvious 
enough in many specimens, including some among the Mysore material. 
Woman’s (1949) trapezoidal and rectangular sections may in part be due 
to joint-planes. In the flaking stage, too, attention had to be paid to the 
future act of grinding. Very often, rough-outs were prepared by coarse 
flaking, and the edges and eminences were then worked off by controlled 
flaking and sometimes retouching of the edges. This had the purpose 
of straightening the edges before the grinding began and was occasionally 
excuted with almost “ Acheulian ” care. 

For certain shapes, special lumps of rocks were required. Woman 
(1949) points out for instance that rectangular, flattish celts were made 
from large flakes taken from cores from which flakes had been removed 
previously, so that both Surfaces of the flake were parallel. 

Interesting evidence of grinding, particularly of the cutting edges was 
found by Foote (1916) on the North Hill at Bellary. He noticed five or 
six well-polished grooves seven to eight inches long and one to one and a 
half inches deep, on the rock surface. The grooves all lay parallel together and 
in close order within an area of less than 20 inches square. Similar grinding 
grooves were found at Helalagundi, on the 19th February, 1889, in Alur 
Taluk. Grinding grooves which were excellently preserved also occurred 


on the Kappa tralla hill in Pattikonda Taluk, Kurnool District and Pullay- 
yagudda in Hyderabad State respectively. That these grooves are associated 
with the axe-makers is further confirmed by De Terra’s (1942) discovery 
of a large boulder of quartzite with seven grooves, each measuring six 
inches in length and one and a half inches in depth, near Bursahom, another 
neolithic site, in Kas hmir . 

So far, no grinding grooves have been found at Brahmagiri. In view 
of the fact that axes in all stages of manufacture are found at Brahmagiri, 
it is not unlikely that such grinding grooves are also present at that site. 

To sum up, it must be said that the process of manufacture was re- 
markably complex. It cannot be described in two stages as Worman has 
done, and even Foote’s stages do not fill the bill. The complete process 
would have been as follows: — 







selection of natural lump or striking of primary flake, 
coarse flaking (rough-out), 

retouching by further flaking and straightening of edge. 




| alternating repeatedly 


Apart from the numerous axes picked up from the surface, Brahmagiri 
has yielded specimens in situ during the excavations in 1940, 1942, and 
1947. The 1947 excavations alone yielded 15 complete specimens and 29 
broken ones. All the axes, it is interesting to note, belong to the pointed- 
butt type. Besides the large number of celts, the presence of cores and 
flakes from the occupation-strata clearly shows that Brahmagiri was a factory- 
site, a centre of local industry. From a close study of the material, two 
stratigraphical observations emerge: — 

(1) The majority of the axes belongs to phase IB. 

(2) A particular type, namely the flattened-lenticular type, is con- 
fined to the ‘ A ’ phase. Only a single specimen (Br. 22-333) of this type, 
however, has been found so far in situ. It is a small flat axe with roughly 
parallel sides which, at the butt end, converge abruptly. An identical speci- 
men, flat and thin, was picked up from the surface by the writer (PI. XIX, 
fig. 3). Similar types were found beyond the Mysore border at Sanganakallu 
(Subba Rao, plate XXI, 9, 1948). De Terra (1942) recovered a rectangular- 
sectioned celt from the bottom level of his trench 2 at Bursahom. This 
evidence is at present taken to indicate that celts of this type (with rectangular 
section) characterise the early series. It should be clear, however, that the 
evidence is still too scanty to be sure of this. At Brahmagiri, a solitary 
in situ specimen is known, whilst at Sanganakallu the horizon of these celts 
is unknown. It Kashmir they were indeed found at the base of the series, 
but this part of India is far removed from Mysore. These flat types were 
probably adze-blades, and there is no reason why they should not be present 
throughout the neolithic period. It should not be assumed, therefore, that 
their restriction to the lower level is a securely established fact. 


In the excavations report (Wheeler, 1948, p.247) the Brahmagiri axes 
are classified according to shape. In doing so, Wheeler found it convenient 
to subdivide the material into a group A in the flat or flattened-lenticular 
section, and B, with lendcular or ovoid section. Each group is further 
divided into three sub-groups and these sometimes into minor variants. 
Whilst this scheme serves well the purpose for which it was intended, namely 
to describe the shapes, comparison with material from several other sites 
has brought the writer to the conclusion that there is little stratigraphical 
or even functional significance in these types, except for the fundamental 
division into axes, adzes and chisels already made by Wheeler. 

There is an indubitable adze, from the upper level of the B culture, 
which is figured in the excavation report, fig. 33, p.249. A further specimen, 
made on thin flake, was found by the writer (PI. XIX, fig. i). 

In addition, there are further thin specimens with symmetrical cross- 
sections of the cutting-edges, very variable in shape and in the length and 
straightness of the cutting edge. To this group belongs the specimen 
Br. 22-333, already discussed, which has remarkably straight edges. It 
comes from level IA. A similar specimen is Br. 21-277, from the middle 
stratum of the IB culture. Wheeler rightly regards it as a variant of IA, 
and no stratigraphical significance can be attached to it. Functionally, 
they must have been different from the heavy, pointed-butt axes. Perhaps 
they were adzes, or else they were used as axes or adzes as the occasion 
demanded. Polynesian tribes make such adze-axes, with a tumable sleeve. 
For this reason, this group is tentatively called here “ adze-axe blades.” It 
is characterised by its thinness, its manufacture from the thin flake, and 
its symmetrical edge. Shape and length of the cutting edge are variable, 
as is the outline of the whole specimen. 

Though all good pointed-butt axes of Brahmagiri come from level IB, 
the broken specimens from IA are evidently like them. Otherwise they 
would have been described in the excavation report. Furthermore, Subba 
Rao did not establish a sequence based on axe-evolution at Sanganakallu 
Thus, the material so far available is still insufficient to distinguish typologi- 
cal phases by means of the polished stone axes. This does not, of course, 
preclude the possibility that future work might reveal such sequence. But 
the writer does not consider the prospects as bright, in the view of the fact 
that the polished axes (and adze-axe blades) so lar lound elsewhere, mostly 
conform to the Brahmagiri type, provided the “ Burmese type ” is regarded 
as a group apart. 

To sum up, the following types from Bramagiri may be distinguished, 
namely: — 

(1) The pointed-butt specimens. 

(2) Adzes. 

(3) The axe-adzes. 

(4) Chisels. 


In discussing the use of the polished stone axes, two alternatives suggest 
themselves. It is possible that the axes were used loose, in the hand. It 
seems more likely, however, and is in agreement with evidence obtained in 


Europe, that they were suitably hafted. That this is the usual practice 
regarding polished stone axes is shown by the cultures of certain tribes of 
Australia and America. In addition, there is some archaeological evidence 
from Asia. Mallowan (1935) in his excavations at Tall Arpachiyah (his layer 
TT6) discovered a wooden haft with a curved handle. A stone celt was 
found lying on the door of a house. The haft was reduced to a thin film 
of carbonised wood which survived in the form of powder, but the outline 
was still clearly preserved in the soil. 

Another interesting discovery is an axe with a groove at the wide butt- 
end, made in the Banda District of the United Provinces, India. This, 
however, appears to be the only grooved specimen known from India, 
and therefore should not be taken as a type characteristic of the Stone Axe 
Culture as a whole. It is an exception and perhaps of very recent age. 
Being broadest at the butt end, it differs radically from the pointed axes 
known from elsewhere. Grooved hammerstones, however, are common. 
The axes found at Brahmagiri and Bellary afford no direct evidence of 

Foote (1916) suggested a method of hafting which he called ‘ spearhead ’ 
type. A bamboo would be cut about three to three and a half inches above 
one of the joints to fit over the butt-end. The implement thus encased in 
the socket would be fastened by a strong lashing. This method cannot 
be conveniently applied to axes used purely for hacking or chopping pur- 
poses, since the handle is parallel to the tong axis of the implement. 

Coghlan (1943) has illustrated several methods of hafting axes and 
adzes. It is obvious that all these methods cannot be applied to the Brahma- 
giri assemblage of celts. His suggestion that the form of the axe largely 
conditions the mode of hafting must be borne in mind while deciding the 
mode of hafting used by the makers of the polished stone celts of Brahmagiri. 

It has been already pointed out that the majority of axes from Brahmagiri 
and other parts of South India are pointed-butt specimens. They are 
thick-bodied and heavy, and thereiore the following methoas of 
Coghlan, namely “ club ” haft (his fig. 16), “ slot ” haft (his fig. 12), 
and modified “ slot ” haft are the only ones that could have been used. 
Such hafts can accommodate heavy and powerful axes, such as those of a 
length of eleven and a quarter inches and a width of cutting edge of three 
and a half inches. Such a specimen was actually found in 1947 on the 
surface at Brahmagiri (preserved in the museum of the Department of 
Archaeology, Mysore). 

The small thin adze from Brahmagiri might have been mounted on a 
“knee” shaft, as illustrated by Coghlan (his fag. 11, p.37), and other aazes 
might have been mounted the same way. 

But there are many ways of mounting adzes, so that it is useless to 
speculate. It should only be noted that the thick adze described by Wheeler 
from Brahmagiri has a pointed butt, which suggests a haft of the ordinary 
hole-type, unless a sleeve was used. Australian natives achieve all manner 
of hafting by using resin in quantity. The possibilities are many, therefore, 
and certainly more varied than suggested by Foote. 

The “ Burmese ” type axes and adzes with their straight butt-ends may 
have been mounted on step-hafts, a rectangular step providing a stop for 
the blade, which prevented it from being driven further into the haft. 


One more word should be said about the adze-axe blades. Though 
they may have been hafted in many different ways, they would lend them- 
selves to mounting in pivot-sleeves. Such mounts are known from certain 
Melanesian islands, where a club-shaft with a large hole holds a long sleeve- 
pieces with the blade at the end. This sleeve-piece is thickest near the 
made, but its thin end protrudes from the other side of the shaft-hole. The 
sleeve, therefore, can be turned at will, so as to use the implement either 
as axe or as adze. That this ingenious contraption may not be restricted 
to Melanesia is suggested by a neolithic specimen in the Landesmuseum, 
Zurich, figured by Clark (1952, pi. IX, fig. b.). 


The dating of the polished Stone Axe Culture of India is complicated by 
the survival of the polished axe into the Iron Age. Worman’s (1949) 
suggestion that this culture lingered on in South India until the 4th century 
B.C. is quite consistent with the Brahmagiri evidence, provided one accepts 
the few specimens found in the Megalithic levels (7) and (8) as contemporary 
with the Megalithic. It is on this evidence that Wheeler holds that there 
is an actual overlap between the polished Stone Axe and the Megalithic 
Cultures. Pottery evidence can be adduced to strengthen this view. If 
it is right, the polished Stone Axe Culture would have existed in a pure 
condition at Brahmagiri in the 4th century B.C. and perhaps still in the 
third. By the second century the Megalithic builders would have appeared 
and mixed with the stone axe makers. This argument has, however, the 
weakness that the specimens occurring in the overlap levels cannot be 
proved to have been contemporary, even if the axes which appeared in 
die Andhra layers are discounted as strays. 

What is the date of the beginnings of the polished Stone Axe Culture 
in the Indian subcontinent? The discovery of stone axes at Copper and 
Bronze Age sites is significant. Marshall (1931), Mackay (1937-38), Har- 
greaves (1929) and Ross (1946) found them in Mohenjo-daro (Pakistan), 
Nal in Baluchistan and Ranaghundai. But all these are stray finds, outside 
a stratigraphical context. Assuming that they are indeed contemporaneous 
with the Indus Valley civilization, a reasonable date would be of the order 
of 2500-1500 B.C. The possibility cannot be ruled out, however, that on 
all these datable sites, the Bronze Age was preceded by a pure neolithic 
stone age complex. How many centuries should be assigned to this stone 
age complex can only be guesswork. It would not be unreasonable to assign 
4 to 5 centuries for the development of such pure “ neolithic ” stone age 
culture, taking its inception back to perhaps 3000 B.C. In fact, Worman 
(1949) dates the appearance of the Indian “ neolithic ” culture between 
3000-2500 B.C., a view which is in agreement with the suggestion made here. 

The next problem is the dating of the beginnings of the Stone Age 
Culture of Brahmagiri in particular. Paterson who in 1936 made a study of 
Foote’s collection of Bellary “ neoliths ” assigned an enormous antiquity 
to them. (1) Though typologically the celts of Brahmagiri have a striking 

(l) Id Ajxclent India, No. 3, p 39, Knshnaswami relates this view of Paterson 1 have been unable to 
find a paper by Paterson himself explaining this matter 


similarity to those of Bellary, the time-relation between these industries 
should not be based on typology but be solved stratigraphically by a large- 
scale excavation. 

The 1947 excavations at Brahmagiri, carried out by R. E. M. Wheeler, 
have already provided some evidence lor the solution of this problem. The 
sequence of the various cultures is, according to Wheeler: — 

III. ‘ Andhra ’ Culture: c. 1st cent. B.C. to 1st cent. A.D. 

II. Megalithic Culture: c. 300 B.C. to 1st cent. A.D. 

I. Stone Axe Culture: c. 1000 B.C. to 300 B.C. 

. Ill and II are datable from historical finds, the Andhra coins being 
associated with Roman coins of 1st century B.C. and the typical pottery 
from other sites providing the evidence for dating III. The Megalithic 
culture is dated by Wheeler (1948, p.202) as beginning after Asoka. The 
deposits of IA and IB cultures are 9 (l > feet thick. Wheeler (1948, p.201) 
has suggested a mean accumulation rate of 3 to 4 feet of occupation soil 
in two centuries. He is very cautious in proposing this figure, which is 
based on a method of calculation of time for deposits applicable to Sirkap 
and Bhirmound of Taxila. If one applies this rate to the IA and IB 
cultures of Brahmagiri one obtains a duration of at least 600 years. Con- 
sidering that the buildings of the polished Stone Axe Culture are likely to 
have been of a flimsy nature, the accumulation rate may well have been 
slower. Accepting the minimum of 200 B.C. for the Megalithic Cutlure, 
the minimum date of 800 B.C. is obtained for the beginnings of the polished 
Stone Axe Culture at Brahmagiri. It may, however, well have been earlier. 

It must be admitted that these figures are too vague to be satisfactory. 
Being what they are, they suggest that the polished Stone Axe Culture of 
South India was later than the polished axes of northwest India. This 
must on no account be regarded as the result of sound investigation, but 
merely as a suggestion, which should stimulate further research. 

(1) In the cutting, Brahmagiri 21, there occurred a sterile layer (18) on layer (19) Similar evidence was 
observed in Brahmagiri (17) The significance of thw level is not yet known Whether it represents a 
break in the occupation, with temporary desertion of the sue remains to be determined Tins is certainly 
possible since, according to Wheeler (1948, p 22) the decorated and incised wares stop at the sterile layer, 
soil samples from sections Br 21 and 17 have been obtained They will be investigated in due courtc. 


Chapter V 



The Mysore area is remarkable for its megahthic remains. In this 
respect it resembles Bellary, Hyderabad, Malabar, North Arcot and other 
parts of the Madras Presidency, where large numbers of megaliths have been 
discovered. In fact megalithic monuments are even more numerous than 
mediaeval temples, for which Mysore is equally famous. Sometimes both 
occur side by side, a fact which illustrates die control that the raw material 
exercises on the location of such activities as temple building and megalithic 

More than 35 megalithic sites have been discovered (PL V) and one 
of them, namely Brahmagiri, has been scientifically excavated. Generally 
they are located at the foot of granite hills overlooking irrigated fields. 
Brahmagiri, Gallapalli, Mallapuram, Addaguppa are good examples. Hunt 
(1924) noticed the same fact in Hyderabad, as did Kristinaswami in his recent 
megalithic explorations of the Chingieput area. 

This location appears to be the result of the megalith builders’ depen- 
dence on a suitable building material. Crystalline rocks have, in India, 
certain qualities which make them eminently suitable for the purpose of 
building, qualities which are due to their rectangular jointing coupled with 
thermal disintegration and, to a minor extent, chemical weathering. These 
rocks break easily into large slabs; they are concentrated as boulders and 
slabs on hillsides and they are easy to work with iron tools (see Newbold 

The granite slabs, which form the cists, could easily and conveniently 
be detached from the nearest part of the hillside by fire and weagcs. un :ss 
indeed they had already been prepared by nature. The granite boulders 
used for the stone-circles could be roughly shaped where they lay and 
rolled down before being carried or pulled to the place of erection by men 
or bullocks. The occurrence of suitable raw material quite close to the 
actual site of the megalithic monuments thus saved a good deal of labour 
and transportation. It is this combination of factors, which largely explains 
why South India and the non-trappean Deccan constitute a classic land of 
megaliths. The megalith peoples might have migrated elsewhere, but 
been unable to build megalithic monuments where the rock did not lend 
itself to the method they employed. 


A note on the geographical location of the megaliths, district by district, 
in the Mysore State is herewith provided. 

Mandya District At one place only, namely Halagur, 12 miles from Mala- 
valli, a Taluk town, a few stone circles have been observed. 


Mysore District. So far, this district has not yielded many megalithic 
sites. Foote collected megalithic pottery from many places on the bank of 
the Cauvery river. But today no signs of megaliths are to be found there. 
However the isolated conical hill of Bettadapur which rises 4,389 feet above 
sea level, has at its foot a number of stone-circles. The village there is 
situated on the Periyapatna-Hassan road. 

Kadur District. At present two sites are known. 

The village of Mattavara is situated in the Chikmaglur Taluk. It has 
attracted attention because of its mediaeval Jaina temple. Equally important 
is the large number of cists without stone-circles on the Government planta- 
tion to the southwest of the village. Most of the slabs of the cists have 
unfortunately been pulled out for building the Taluk office at Chikmaglur. 
The slabs measure about 9 by 8 feet by 10 inches. Most of the cists have 
been opened. Some spears and pottery, are said to be in the charge of 
the local police. 

Sakkarepatna, a small town 14 miles northeast of Chikmaglur on the 
Kadur-Chikmaglur road is surrounded by the famous Bababudan mountains 
and harbours a few stone circles. 

Bangalore District. Bellandur, a village near Bangalore, has some cist- 
circles. Instead of rough boulders, rough slabs of various sizes were used 
for the circles, standing in a slanting position and deeply buried. The site 
is on the property of Mr. E. M. Morrel of Bangalore. Mr. Bush excavated 
four of them and collected some pottery and iron objects. The fifth, exca- 
vated by the Archaeological Survey of Mysore in 1916, revealed a cist with 
a port-hole. 

In the vicinity of Channapatna, 37 miles southwest of Bangalore, on the 
Bangalore-Mysore road, some cist-circles are to be seen. At Banvali, one 
mile from Channapatna a megalithic tomb was excavated by Mr. Kandoji 
Rao on the 12th September, 1897. A thick-walled and red polished pot 
(broken) with four thick legs, together with some iron implements and a 
bronze ear ornament from this grave, are in the Madras Museum.*') 

Stone-circles of varying sizes are seen near the village of Managondahalli, 
8 miles from the Taluk town of Devanahalli. Two of the stone-circles 
were opened <2) and contained no cist in either case. Previously another 
stone-circle had been opened by the headman of the village, who had collected 
some interesting pottery. It had a cist, which suggests that both pit-circles 
and cist-circles exist here side by side. 

Savandurga, 22 miles west-southwest of Bangalore, is an enormous mass 
of granite, in the Magadi Taluk, which rises to 4,024 feet above sea level. 
At its foot, a large cist-cemetery lies in the jungle. Apart from the cists and 
stone-circles, the site contains other types of megalithic monuments. There 
is a fine dolmen, and examples of free-standing chambers (i.e. almost entirely 
above ground). 

The cists are oblong and covered with capstones, which are huge un- 
dressed slabs 8 to 14 feet in length, 7 to 10 feet in width and 6 to 16 inches 

(1) Numbers 263-1 lo 263-9 of Catalogue Raisonoc of Foote, 1914 

(2) Report of the Archaeological Department, Mysore, 1917 No sections, plans or any other evidence of 
stratification are available 


in thickness. The cist-slabs are 5 to 10 feet in length, 4 feet 6 inches to 
5 feet 6 inches in height and 2 inches to 6 inches in thickness In view of 
the comparative thinness of the cist-slabs supporting the huge capstone, it 
is not unlikely that the cist was surrounded by and covered over with earth 
before the capstone was placed on top. At this place were found double 
or treble stone-circles. 

Near the village of Koramangala, Magadi Taluk, there is a field of cist- 
circles with rough boulders and gigantic slabs. 

The village of Sonnappanhalli is situated 131 miles north of Bangalore. 
Its cist-cemetery shows port-boles 21 feet in diameter. 

A vast cist-cemetery, covering an area of more than one mile occurs 
to the east of the village of Chikjala, IS miles north of Bangalore. The 
cover slabs are 15 feet by 15 feet and in two cases where the cist is visible 
the eastern slab has the port-hole. Pottery and iron implements collected 
from the tombs by Col. Branfill are in the Madras Museum. 

Hassan District. The Hassan District has a large number of megalithic 
sites. It attracted the attention of Captain J. S. F. MacKenzie as early as 
1873. Certain mounds, however, which he saw above the ground are 
perhaps not megaliths, since their edges are not marked by the usual stone- 
circles. The stone structures which he described are small. Today, they 
are used as temples for the village goddess. Though they may have 
originated as prehistoric monuments, the possibility of their recent age must 
be considered seriously. Wheeler, for instance, figured a Naga shine from 
Brahmagiri, which is quite modern, and Elwin (1951) refers to the various 
types of monuments to the dead originally or even now made of stone by the 
Bhil of central India. D. H. Gordon further believes that the wooden posts 
of the Korku are derived from stone prototypes. In short the tradition of 
using stone for the making of monuments is by no means extinct in India, 
and it is indeed probable that Mackenzie’s stone shrines belong to the 
“ recent ” category. Mackenzie also noted a few menhirs and stone-circles. 

A recent exploration of this area has yielded a further number of stone- 
circles, such as Gunni and Halebid, two out-of-the-way villages. Honnavara 
and Pumgame are villages, 5 miles from Dudda, a small railway station on 
the Mysore-Hassan line. The megalithic monuments at Pumgame stand 
on high ground 200 yards to the south of the village. Here and there 
menhirs are seen. 

Kondajji is a small village about 6 miles to the northwest of Saligame. 
About a furlong from the local temple, there are to be seen on ris ; ng ground 
two dozen stone-circles, with an average diameter of 25 feet. On an emi- 
nence to the north of die village of Sompur in Arkalgud Taluk, a number 
of stone circles enclose small mounds. Some of them appear to have a pit 
or cavity in the centre and therefore may be pit-circles. 

Ttunkur District Only one site is known from this area. A low range of 
hills begms at Kibbanahalli, and runs along northwest of Chikkanayakana- 
halli. On the crest of the low ridge 3 furlongs west-southwest of Kerala 
Katte, H. K. Slater (1906) noticed half-a-dozen cist circles. The cists are 
8 feet square and composed of granite slabs. They are obviously in a dis- 
turbed condition, and it has indeed been reported that they were rifled a 
few years ago. A few more stone-circles two furlongs southeast of point 
2203 are fortunately intact. 


Kolar District. This District abounds in megalithic sites. Just south of 
Garudanahalli, 2 miles north of Narasapur in me Malur Taluk, lies a range 
of gneiss and granite hills, extending to the Kolar hills. Near the village, 
covering an area of half a square mile, are about 100 dst-circles (Cole, 1873). 
The port-holes at this site are not more than a few inches in diameter and 
covered by a round slab. Some menhirs are also found near the stone- 
circles. Mr. Cooke of the Kolar Gold Fields opened some of the tombs in 

Numerous cists occur in the valley between Avati hill and Koligudda. 
The slabs that cover these are very large and measure 9 to 10 feet square 
with a thickness of 1^ feet to 2 feet. In Arthur Cole’s time a few of these 
top-slabs were unfortunately blasted to pieces to remove the grave goods. 

On an eminence to the east of the village of Garudanapalya, there are 
a few cists. The landlord of the place found it profitable to remove most 
of the slabs in order to build a well. Some of the finds are reported to be 
in his possession. 

Parandapalli is situated at a distance of 2 miles from the travellers’ 
bungalow of Robertsonpet. This village lies in the midst of a number of 
rocky heights. To the east of the village, about two miles away (near the 
7/4 furlong stone from Betamangala) on the top of a rocky ridge and on its 
sides, are seen more than 100 stone-circles, appearing just above ground 
level. They have a diameter which varies from 7 feet to 40 feet. 

On the Kolar-Betamangala road in craggy country lies the village of 
Hunugundapatna. The locality is given its character by the conspicuous tor 
with a smaller boulder perched on it. The place abounds with cave- 
shelters. An interesting series of pottery, red-and-black ware, possibly of 
the megalithic period, can be collected from the site. But its sequence 
cannot be worked out without obtaining stratigraphical evidence by means 
of an excavation. Two miles from this place is situated the village of 
Hunugunda, (1) Bowringpet Taluk, with several stone-circles. They have 
diameters which range between 15 and 20 feet. Many of the boulders of 
the stone-circles are missing, having been removed by the villagers. 

On the slope of Murugamale hill, 7 miles from Chintamani, traces of 
megaliths occur. Typical megalithic pottery is found, as also at Haralakote, 
5 miles to the east of Srinivaspur. 

While travelling on duty in the capacity of Inam Commissioner of 
Mysore, Captain Cole saw a menhir at Perisandra, near Chikballapur. The 
adjoining village of Mashalli produced 54 port-holed cist-circles with huge 
unhewn blocks of granite as capstones. 

A few feet off the main Gudibanda road, Cole (1873) saw the outline 
of black pottery in the hard gravel where the road had worn down level with 
the mouths of these vessels, the cist-slabs having been removed probably by 
the Woddars or stone-workers. 

He came across a number of stone-circles at Gudibanda itself. On 
approaching Kolar town, near the third mile, he observed some cist-circles. 
He excavated a few and found pottery incised at the rims. 

(1) It is known ai ' Porkunram * in the tarail inscriptions. 


Two miles from Bowringpet, cn route from Kolar, a few pit-circles were 
found. The presence of a slab with a port-hole at die place also suggests 
the existence of cist-circles. Two miles east of the village of Margal, pit- 
circles and cist-circles also occur together, as is the case at BrahmagirL 

Chi tal drug District The northern part of Mysore (mainly the Chitaldrug 
District) is rich in megalithic sites. It is contiguous with Bellary, Rayadrug 
and Anantapur, where hundreds of megalithic monuments are found. On 
the evidence of pottery produced by a few megalithic tombs, particularly 
Guntakal, all these regions probably belong to one cultural zone. More 
intensive work, however, is required in this area to confirm this suggetion. (1) 

At Nandanhosur, 12 miles west of Chitaldrug, Ganjippata, 6 miles 
west of Chitaldrug, and at Budnapur, 3 miles west of Chitaldrug, stone 
circles occur. 

Foote in 1896 collected some pottery from cists obviously disturbed 
by woddars (stone masons) from the village of Talya in the Holalkere Taluk 
of the Chitaldrug District. 

The Andhra site of Chandravalli is situated at the northwest foot of the 
Chitaldrug hill. There is an extensive valley close to the modem town of 
Chitaldrug. Half-a-dozen cists of small size without stone circles and top 
slabs were found by Dr. Krishna (1929) during trial excavations. Excava- 
tion 5 revealed a small cist 3 feet by 2 feet which appears to have contained 
skeletal remains. 

Near the town of Molakalmuru, a few stone-circles were observed by 
Mr. V. D. Krishnaswami, but the most extensive and important megalithic 
site in the Chitaldrug District is Brahmagiri, a few miles north of Molakal- 
muru. At a distance of half-a-mile east of the remains of the ancient 
Asokan town, there occurs a vast field of megalithic monuments. Many of 
them have been removed by farmers, but a few hundred still survive here 
and there. 

It must be mentioned that the mere presence of the megalithic struc- 
tures may not by itself be of great interest to the archaeologist. Even to 
excavate a hundred of them with the utmost care would not necessarily 
add significantly to our knowledge of their chronology. The crux of the 
problem is to find a site where the megalithic culture can be correlated 
with a dateable culture-sequence. This would enable progress to be made 
from the known to the unknown. Luckily Brahmagiri has provided this 
type of evidence and, to a certain extent, solved the riddle of die me g aliths . 
The occurrence of pottery forms and fabrics, typical of the megalithic tombs 
in the stratified deposits of the Asokan town-site of Br ahmagir i, provided 
the clue to their date. 

The megaliths of Brahmagiri (Wheeler, 1948) fall into two categories, 

(1) cist-circles (Megaliths I, IV, V, VI, VIH, X). 

(2) pit-circles (Megaliths II, III, VII, IX). 

(1) Though there is a common megalithic complex, the megaliths may not belong to one culture The three 
regions subjected to a detailed ground survey since 1944, namely Cochin Pudukottai and the Chingleput 
District, have shown the individuality of each Pudukottai is characterised by the elaborate transepted 
port hole cist Cochin has us typical toptkaU and rock cut caves common on the Malabar coast The 
dolmcnotd cists of Chingleput invariably enclose a terra cotta legged Sarcophagus (Ancient India, No 5, 1949, 
p 36) To this we may add the Rrahm lgin Bellary Rayadrug area wan us cist circles and pH circles 
which forint a separate entity by itself 


In the first type the cist at the centre is surrounded by a stone circle. 
The pit circles, as the name indicates, has a pit instead of the stone-cist. 
The outline of this pit in the centre of the stone-circle can be traced and 
planned by careful excavation. 

Adjoining Brahmagiri but beyond the political boundary of the Mysore 
State, many other megalithic sites occur. Sanganakallu, 3 miles from 
Bellary, is an important site (Subba Rao, 1948). Guntakal, which is a rail- 
way junction, has yielded pottery from megalithic tombs strikingly similar 
to certain pottery forms of Brahmagiri. Kalyandrug in the Anantpur Dis- 
trict is another site.* 1 * Rayadrug, a neighbouring district, has produced 
from its several sites (namely Mallapuram, Addaguppa, and Gallapalli) 1210 
megalithic monuments, out of a total of 2129 from the whole Bellary District. 
More than three-fourths of the total of megalithic monuments are concen- 
trated here. 

The geographical entity formed by the valley of the Chinna Hagari 
river which comprises both Rayadrug and Brahmagiri (see Chapter I, 
and PI. V), shows certain cultural affinities. The whole valley is studded 
with megalithic sites, which indicate that the same people occupied these 
areas and practised the same burial customs and that their culture survived 
into early historic times, and a large number of sites occur with polished 
Stone Axe Culture associated with microliths (see Chapter III). 


The main types of megalithic monuments found in the Mysore area 
are: — 

(1) Dolmens. 

(2) Menhirs. 

(3) Cairns. 

(4) Port-hole cist-circles. 

(5) Pit-circles. 

Making allowances for their own individuality 1 (2) as evidenced in certain 
areas, no one particular type is confined to one particular site in the Mysore 
area. For example, at Brahmagiri, we have the menhirs, pit-circles and 
cist-circles. At Savandurga in the Bangalore District, dolmens occur to- 
gether with cist-circles. In a few sites of the Kolar District, pit-circles 
and cist-circles exist side by side. 

Most megalithic structures are of a complicated design and a brief des- 
cription of their architecture is necessary to make clear the ideas and motives 
relating to the burial customs of the megalithic people. Even a superficial 
examination of the structures makes it clear that a huge effort was involved 
in bringing them into being. Two alternatives emerge. The first is that 
the whole community was responsible for the erection of the monuments, 
sharing the effort incurred. The other is that building a megalithic monu- 
ment with its elaborate grave furniture, such as choice pottery, iron imple- 
ments and in some cases finely made beads of etched camelian and gold, was 

(1) Annual Report of the Archaelogical Department, Southern Circle, Madras 1912-13 (1913), pp.53-62, 

(2) Ancient India, No 5, 1949. p 36 


Kf v* 

•* a 1 

exclusively confined to the richer stratum of society, who could muster the 

There are five stages in the construction of a megalithic tomb, namely: — 

(1) quarrying of the required stones, 

(2) moving of the stones, blocks and boulders to the desired spot, 

(3) erection of the orthostats in their proper places and drystone wall- 
ing round the cist, 

(4) placing of the capstone or cover-slab on the cist, 

(5) placing and arrangement of granite boulders so as to form a stone 
or boulder-circle. 

As already suggested (p.59) quarrying could only be done at a place close 
to the burial site. The stones and the boulders for the circle or circles 
must have been rolled down from the granite hill slopes. The slabs were 
probably obtained by lighting a fire on the surface of the rock with a view 
to loosening the top layer by physical expansion and contraction along a 
horizontal plane and then inserting wedges into the line of cleavage. This 
was not difficult since weathering had prepared the rock for the cleaving 

The slabs were set up in such a way that one end of each slab pro- 
jected laterally across the end of one of the adjacent slabs, thus forming a 
sort of swastika, either anti-clockwise or clockwise. This is supposed by 
some to be a device of symbolic significance (Ghurye, 1926), but in fact it is 
hardly more than a constructional necessity. It locked the four sides to- 
gether and prevented an inward collapse of the orthostats. 

Another curious feature is worth mentioning. Each cist formed by 
slabs is surrounded by a dry-stone wall which reaches almost to the level 
of the capstone and, on plan, exhibits a spiral arrangement. 

Where double stone-circles occur, as at Brahmagiri, enclosing a pit or 
a cist, as the case may be, the space between the outer and the inner circles 
is closely packed with rubble. The two details, just mentioned, are of 
course visible only after excavation. 

The port-hole is an invariable accompaniment to the megalithic cist, 
n northern and western Europe it was made by striking out semi-circular 
notches in the edges of two adjacent uprights, but in the Crimea, Caucasia, 
oyna and India it is a circular hole cut in the eastern orthostat. 

It is difficult to explain the occurrence of the port-hole as a feature 
t0 J he mega i thlc clsts of 130111 east and west. “ A hole is a hole 

Sfi SSi-yp as-sLjrS 

structures of Africa, India, Indonesia and Japln have verv £r? in 
with the burial chambers of prehistoric 

S3T 4 ,he ^ 


(1852) that the megalithic remains belonged to the Celtic-Scythians, can no 
longer be seriously maintained. Judged by the contents, the tombs in ques- 
tion do not belong to a single culture and therefore were not erected and 
used by a single people (Childe, 1950a). 

In northwest Europe the port-hole is inseparably connected with the 
threshold or entrance idea (Kendrick, 1925). Whether the Indian examples 
were intended for the introduction of later sepulchral urns when occasion 
required, or for blowing incense through long tubes (Walhouse, 1874), or 
for passing food (Peet, 1912), or providing an outlet for the soul, is a matter 
for speculation. Referring to the port-holes of Brahmagiri, Wheeler (1948) 
suggests that the bones which had to be deposited in the cist were thrown 
through them. 

Daniel (in Clifford and Daniel, 1940) restricts the term port-hole only 
to the European examples because they are fairly large. To those of India 
and Palestine he applies the term ‘ ghost-hole ' because of their small size. 
This classification cannot be maintained in India because fairly large-sized 
port-holes do occur there. One of the cists at Savandurga in the Bangalore 
District, for instance, has a port-hole large enough for a man to pass through. 
Some of the cists in Hyderabad also have large port-holes. 


The grave goods comprise pottery, iron objects, stone objects and beads. 
Since the stone tools, which are of the greatest interest in the present con- 
text, play but a subsidiary role, it is necessary to comment briefly on the 
other material also. 

Pottery . — Foote (1914 and 1916) refers to the megalithic pottery vaguely as 
' Iron Age ’ pottery. Broadly it falls into two categories, coarse unpolished 
ware, which largely consists of urns and storage jars, and fine polished ware. 
A cist at the village of Managondahalli produced a large storage jar, 3 feet 
in height with an ornamental band at the neck. ,l) Some grain was actually 
collected in an excavated cist-circle in Mysore from such pottery vessels (Bain, 
1890). The thick elephantine-legged, coarse, unpolished vessel found in the 
megalithic tombs is, of course, a burial urn 

The fine polished ware is often red and black; some pots are entirely 
black. The red-and-black effect was perhaps not intentional, being the result 
of unequal firing. The exposed part of the pot acquired the red colour 
owing to oxidation of the iron contained in the clay. Generally the lower 
portion of the vessel is red, while the upper portion is black, owing to 
contact with the reducing atmosphere produced by the burning fuel, the 
vessel being inverted during the firing. The black polished ware, distinctive 
of the pit-circles, was produced by low temperature firing and therefore 
weathers much more easily than the red. 

It is interesting to note that the pottery types of the pit-circles and of 
the cist-circles are somewhat different. Lids with many sub-types, as op- 
posed to small knobbed lids of the cists, appear to be more common in the 
pit-circles. Those found by Wheeler at Brahmagiri have interesting parallels 

fl) Annual Report of the Archaeological Department, Mysore, 1917 


from other sites. No. 1286 of Foote from Chikjala in the Bangalore District 
is almost a replica of Wheeler’s P.4 (1948, fig. 13). Nos. 2605d and 2605c 
(Foote, 1916, PI. 66) from the Cache at Patpad, Kurnool District, are mere 
variants of Wheeler’s P.4 and P.5 No, 1231 from Guntakal (Foote, 1901, 
PI. XXXIII) is a variant of Wheeler’s P.1, and 

Some interesting specimens of pottery were recovered from the cists at 
Savandurga and Chikjala in the Bangalore District. Two specimens re- 
semble a horn or gourd and are hollow (see Raghavan, 1935). Bruce Foote 
(1901) suggested that they were used as seed boxes, an unlikely interpre- 
tation. 'This type is unknown from Brahmagiri. 

The thick-rimmed and wide-mouthed pots and pans from Huttanhalli, 
18 miles north of Bangalore, are somewhat similar to these of Brahmagiri. 

Footed vessels are one of the most widely distributed pottery types from 
the megalithic tombs of South India. The localities from which they have 
been recovered in Mysore State are as follows (distribution map, PI. V): — 
Brahmagiri, Chandravalli, Talya, Managondahalli, Chikjala, Savandurga, 
Bellandur, Garudanahalli, Huttanhalli. (l > 

The megalithic tombs also contain miniature pottery. Fergusson (1872) 
believed that the presence of miniature pottery was an indication of the 
pygmy stature of die race of people who were responsible for the burials. 
The miniature vessels are not, however, found in miniature tombs but in 
megalithic tombs and structures of normal size (Walhouse, 1875). Besides, 
they are found side by side with large-sized pottery. They form a regular 
part of pave furniture and contain in all probability, some offering. It 
is possible, for instance, thait they were perfume containers or scent bottles 
for the dead. <2) The miniature cups Nos. 21 and 23 from Savandurga cists 
resemble Odagattur No. 317 — 1923. 

Graffiti on megalithic pottery are common* 3 ' but they occur also on urn 
burial pottery (Laffitte, 1931). These marks have been differently interpreted 
by different writers. Foote (1916) and Richards (1924a) are of the opinion 
that they represent ownership marks. That they were the personal marks 
of the dead cannot be accepted because the same marks are found on potteTy 
widely separated by distance. For instance, Dubreuil saw on the burial 
pottery at Calicut “ mark 14 ” of the Pondicherry urns. Yazdani (1917) 
who classified 131 different forms of these marks put forward the theory 
that they were the sources of the Brahmi script. It seems on the other hand 
that they are merely post-firing scratchings recorded by Wheeler (1938). 

Iron objects : — The megalithic cists and pit-circles in Mysore contain, as part 
of the grave goods, a large number of iron objects. The axes, chisels, sickles, 
swords and arrowheads from Jala and Savandurga in the Bangalore District 
have been catalogued by Foote (1901). Mr. Bain (1890) who opened 3 
stone circles in the Mysore country (the name of the site is not specified) 
found in the cist two swords crossed upon each other. Swords in a similar 

(1) Annual Beport of the Archaeological Survey of India. 1950-14. PI CXX1X, 1-8 A imall 4-footcd bowl 
wih a ratling ornamentation at the neck was found 


Noi. 1105, 1307 and 1311 (Poore, 1901). 

No. 313—1923 (Indian Section uf the Victorta and Albert Museum) u a black burnished miniature vase 
from Odagattur 


Hum <1916 and 1924) found 
observed some of them on the 
Arcof District Aiyappan (1933) 
of Ferjke, southern Malabar 

them on the Hyderab id megalithic pottery. F ] Richards (1924.V) 
pottery which he excavated from three stone circles at Odagattur, North 
noticed some of the marks on the pottery of the rock cut cave tomb* 


position were found in the megalithic tombs at Guntakal. Since the skeletal 
remains were found by Bain beneath the crossed swords, some kind of 
ceremony is suggested. At Savandurga Col. Branfill (1881a) found iron 
arrowheads stuck into the ground in an upright position, which again implies 
the performance of some rite. Wanchope opened a few megalithic tombs 
at Hutranhalli, which yielded iron celts and spearheads. A celt 10} inches 
long, 5} inches (maximum) broad, has a pointed projection on each side 
just above the cutting edge. It resembles the copper celts of Gungeria 
in the United Provinces (Mazumdar, 1936). 

The cists and pit-circles of Brahmagiri yielded, apart from 20 frag- 
mentary and much-decayed objects, 7 tanged daggers, 4 wedge-shaped blades, 
4 chisels, 2 sickles, 2 lances, a sword, a barbed arrowhead, a chopper-like 
object with a long handle and 3 spears 5} - 6} ft. long. From the dsts 
came 6 tanged daggers, an axe, a fragmentary chisel, besides a few unidenti- 
fied fragments. (1) 

This abundant association of iron with the megaliths of Brahmagiri is 
regarded as being connected with a wave of invaders who came to northern 
Mysore when the Maurya Empire was breaking up (Wheeler, 1948). If 
this invasion-theory is correct, the megalithic culture may have spread to 
central and eastern Mysore from the north. This idea is supported by the 
magalithic culture of Chandravalli which is chronologically later than that 
of Brahmagiri, as is indicated by the discovery of Andhra coins near the 
megalithic burials (Krishna, 1929). Chandravalli may therefore represent 
a further step or stage in the southward expansion of the megalithic folk. 
This evidence again supports the view that the megalithic culture was a period 
of considerable duration, and that its sites should not be regarded as con- 
temporaneous everywhere. 

Beads : — Beads are commonly found in both megalithic and urn burials of 
South India/ 2 ' Thus Col. Tucker found carnelian beads in a cist at Sulur, 
7 miles east of Podunur in the Palladam Taluk of the Coimbatore District. 
Many beads were recovered from the cists and pit-circles at Brahmagiri. 
The megaliths IV and VIII yielded magnesite beads of various shapes. Mag- 
alith VI and pit-circle No. 2 contained terracotta beds. Megalith IX was 
remarkable since from it were recovered, in addition to 33 gold beads, two 
made of steatite and serpentine respectively. 

The stone raw materials used for bead making in the Megalithic phase 
of Mysore were carnelian, magnesite, steatite and serpentine. Of these 
only the first is a very hard substance, and the last two are soft and easy 
to grind. 

But metal and terracotta beads are already present. Evidently, the 
art of stone-polishing was beginning to suffer from competition by the metal 
worker and the potter. The substitution of more easily produced articles 
was going ahead in the megalithic period, and stone was being replaced. 
Artifacts ( other than beads :) — Rea (1912; 1915) found a polished stone celt (3) 
in one of the pottery cists from a stone circle near the reserved forest area 

(1) Nos 1272, 1273, 1274, 1275 and 1276 of Foote (1901) refer to iron Implements from the megaliths 
of Chikjala, and Nos 1330 to 1349 to those from the cut-circles of Savandurga in the Bangalore. District, 

(2) Indian number of ‘ Man ’ (1930) 

(3) See plate XXXIII, No 24 (1912) It measures 4 ms x lj ms and has a pointed-butt. 


of Perambair, Chingleput District of Madras Presidency. Numa Lafitte 
(1931) found stone axes in an urn-field near Pondicherry. Cammiade ( i 924) 
collected a few microliths from burial urns in the region of the Godavari. 
Two small crystal scrapers were found by Branfill (1881a) in a megalithic 
cist at Savandurga in the Mysore State. Apart from these few examples of 
stone implements, the stone objects of the megalithic burials are usually 
confined to mortars and pesdes. The granite pestle from the pit-circle 
at Brahmagiri (Megalith II, No. 12, PI. CXVI, Wheeler, 1948) resembles 
the well-polished diorite pesde which is nearly 12 inches long (No. 324-1923 
I.M.) and which comes from Odagattur in the Madras Presidency and is 
in the Indian section of the Victoria and Albert Museum, London, with 
other objects from Chagattur. Similar stone objects have been obtained 
from the megalithic burials of Jala in the Mysore State. No. 1278 and 
No. 1280 of Foote’s Plate XXI (1901) are a pestle of hornblende and a 
mortar of the same material, respectively. 

Of these stone artifacts, the polished celts are the most important, 
for they provide a link with the preceding “ neolithic ” culture. They are 
exactly ot the same pointed-butt type as in this culture, and very few have 
been found. One might, therefore, hold the view that these are stray 
specimens which, on megalithic sites superimposed on neolithic sites, in'ruded 
accidentally into the megalithic level. But in spite of the scarcity of such 
spec.mens this view is nardly tenable, for one oi them was found imiae 
a megalithic urn near Pondicherry and another inside a pottery cist at 

There is no doubt, therefore, that these axes were used by megalithic 
people. It is a very different question, however, whether they were made 
by them It :s quite conceit able the: thc„e arc neolithic specimen., picked 
up by the megalithic people and to which some magical value was attached. 
This is indeed the case in many parrs of the world today, where polished 
stone celts are regarded as thunderboi s oi other missiles of thw god-. And 
in Mysore and other parts of southern India there aie enough polished celts 
lying on the surface to attract attention 

The stone industry of the megalithic builders, therefore, may no longer 
have included the making of celts Microliths are likely to have continued 
on a small scale, as they have m other iron-using communities, for instance 
in South or East Africa. Stone remained the cheaper raw materia! long 
after the introduction of iron 

On the whole, however, one notices that the megalithic people specialised 
in working stone in the following cases: — 

(a) beads continued to be made. 

(b) mortars, etc., appear, which take advantage of properties of stone 
that are lacking in the metals. 

(c) architectural use is made of stone in the erection of the megaliths. 

The megalithic civilization was thus drifting away from the use of stone 

for weapons and tools. It discovered new uses for stone, above all in the 
introduction of architecture, which in the course of the following centuries 
was destined to become the chief manner of using this raw material in 
southern India. This trend was, of course, the direct result of the develop- 
ment of iron implements. These were on the whole more efficient than 
stone tools, and they enabled man to work stone on a large scale for archi- 
tectural purposes. 



The theories concerning the origin of the megalithic culture have a 
bearing on the chronology of the megaliths. Whether it was basically western 
as Penfca, Kossinna, S. Reinach and Leeds believed or was an eastern idea 
as S. Muller and Montelius thought, need not be discussed here. Elliot 
Smith (1913) considered that the Egyptian mastaba was the parent of the 
dolmen. Peake (1916) suggested an Aegean origin for them. Perry’s 
theory (1915) that the megalithic monuments were the work of the Egyptian 
traders, who went in search of pearls, gold and copper is not tenable because 
the distribution of the megaliths does not always coincide with the areas 
of distribution of those commodities. 

Hunt (1924) revived the question of Egyptian affinities by equating some 
pottery marks with Egyptian ‘ K Walhouse (1874) on the other hand 
pointed to affinities with Etruria. These views only show how inconclusive 
the evidence still is. One is not entitled even to assume a common origin 
for all megalithic cultures. Those of Europe and North Africa may belong 
to one complex, and those of India to another. In any case, as will be 
shown in the following paragraphs, the Indian megaliths are about 1500 
years younger than those of Europe, a difference which does not strengthen 
the hypothesis of common origin. 

Indian legends connected with the megaliths are chronologically mis- 
leading, implying an unduly early date. For instance, most of the South 
Indian megaliths are called ' Pandukulis.’ They are associated with the 
Pandava brothers of the Mahabharata. There is no evidence to show that 
the megalithic burial custom originated with the Pandavas or was in vogue 
during the period of their regime. 

At one time it was the fashion to assign the Indian megaliths to a 
prehistoric period. Some writers even now give them an early date. Ref- 
erences in the Rigveda have led Panchamukhi (1946) to suggest a date be- 
tween 1200-1000 B.C Dubreuil was of the opinion that the rock-cut tombs 
of Malabar with their hollow cave of hcmisperical shape and a central open- 
ing like a chimney were survivals of a Vedic fire-altar. The Vedic evidence, 
however, needs far more careful examination but archaeological evidence 
points to the conclusion that most of the megaliths in South India are assign- 
able to the period between the 4th century B.C. and the early centuries 
of the Christian era. The arguments in favour of this date are as follows: 

1. As has been pointed out on p.67 most of the megaliths are associated 
with a fully developed series of iron weapons and implements: spears, 
swords, daggers, sickles, chisels and axes. Some of this equipment looks 
almost modern. An object of steel without any corrosion was recovered from 
one of the cists at Savandurga in the Bangalore District by Branfill. One 
hesitates to assign to them an early age, a view which is also held by Gordon 

2. Some punch-marked coins have been reported from megalithic 
burials in the Coimbatore area (Wheeler, 1948, p.300, No. 4). Their date 
is known to range, according to Allan, between the 6th and 3rd centuries 
B.C. Most of the other evidence also points to a date later than the 6th 
century, so that this value must be regarded as a terminus post quern. 


3. The problem of the so-called russet-coated Andhra ware deserves 
to be discussed in this connection (see PI. IVB and for a full list of the 
sites of this particular ware see Ancient India (1947) No. 4, p.308). At 
Chandravalli, Brahmapuri, Kondapur and a few other sites, this ware is 
associated with the heavy lead coins of the Andhras and their feudatories 
as well as with the “ Touletted ” ware of Arikamedu. Its date therefore 
has been rightly assigned to the first century A.D., at least in so far as it 
must have been used in that century. It may, however, be earlier or later 

This pottery has certain features in common with the megalithic 
pottery: — 

(1) It is found in association with the usual red-and-black ware. 

(2) In form and shape it corresponds to the red-and-black ware. 

(3) Like the red-and-black pottery it bears post-firing graffiti. 

On a restricted scale it occurs in the megalithic tombs, the Coimbatore 
District being its stronghold. Walhouse (1875) found this type at Nallam- 
patti, 6 miles north-northwest of Perundruai, which lies south-west of Erode; 
Macqueen at Rakiapalayam and Tucker at Sulur (Beck, 1930) — all in the 
Coimbatore District. Foote (1901) also found it in the Coimbatore, Trav- 
ancore and Tinnevelly Districts (Nos. 1076: cup; 1077, 1085, 1086, 1087, 
1088, 1089, 1092, 1120-1123, 1151i, 1151g, 1151h, 1151j, 1151x, 1151y: 
bowls; 1084: vase; 1111 to 1119: pot (Chatty); 1151a: pot (Chatty) and 
1001 bowl). 

The Indian section of the Victoria and Albert Museum has a number 
of specimens of russet-coated ware with wavy lines. No. 5762-1901 is a 
bowl which is red outside and black inside. The red outside has wavy 
horizontal bands in yellowish white (Kaolin). No. 5761-1901 (refers to 
registration numbers of the V. and A. Museum) is a similar carinated vase, 
but with straight parallel horizontal bands. No. 5759-1901 is also a cann- 
ated vase but with wavy design. All these belong to the megalithic graves 
of Coimbatore District. No. 08078 I.S. is a fine carinated bowl with rhe 
wavy bands but its provenance is not known. The label says that it comes 
from a stone-circle in South India, 

Govinda Menon (1937), discovered similar russet-coated ware in a 
megalith at Tiruvilvamala, in Cochin State. Lastly, Dr. Krishna (1942a) 
recovered a vase of this type of pottery with lattice design in kaolin from 
one of the megaliths at Brahmagiri. 

Col. Tucker’s evidence from Sulur, 7 miles east of Podanur in the 
Palladam Taluk, Coimbatore District, is particularly valuable because the 
megalithic cist which contained the russet-coated ware yielded also a bronze 
coin, which was identified by J. Allan of the British Museum as one of 
Eran, of the 3rd century B.C. This evidence suggests that “ Andhra ” 
pottery may have been made as early as the third century B.C., and its 
range was from c. 300 B.C. to c. 100 A.D. 

On Andhra pottery evidence, therefore, the megalithic period would 
have extended into the first century A.D., though the presence of such 
pottery need not always imply so late a date. 


4. The megaliths of Chandravalli, judged by their pottery, appear to 
be different from those of Brahmagiri. They may perhaps correspond to 
the latest megalithic phase of Brahmagiri. Salt-glazing is common at Chand- 
ravalli and absent at Brahmagiri. Wheeler (1948) points out that this glaze 
is frequent in Andhra pottery, indicating a late age, perhaps up to the first 
century A.D. 

Chandravalli therefore appears on the whole to be later than most of 
Brahmagiri, where the Andhra pottery phase is preceded by the “ Mega- 
lkhic ” pottery phase indicating a considerable period of megalithic building 
prior to Chandravalli. 

5. Coin evidence suggests that megaliths were in fact still constructed 
in the first century A.D. At Chandravalli, Dr. Krishna (1929) discovered 
Andhra lead coins of that century. Wheeler (1948) further reports that 
a Coimbatore collector is believed to have found a silver coin of Augustus 
in a megalithic tomb. This doubtful piece of information agrees with the 
remainder of the evide'nce. 

The preceding arguments have been selected as based on material finds 
made in the megaliths themselves. Taken as a whole, they suggest that the 
megalithic period ranges from the 6th century B.C. to the 1st century A.D. 
Whether this will prove to be the latest date remains to be seen, for there 
is some literary evidence suggesting that megaliths continue to be built even 



AIYAPPAN, A., 1933. Rock-cut Cave-tombs of Ferske, South Malabar.— 
Quart J. Mythic Soc. (n.s.) 23(3), pp. 299-314. 

, 1941. Rude stone monuments of the Perumal Hills, 

Kodaikanal. — Quart. J. Mythic Soc., Bangalore 31(3/4). 

, 1945(a). Megalithic monuments of s outh India. — J. Indian 

anthrop. Inst, (n.s.) 1, pp. 31-48. 

, 1945(b), Mesolithic artifacts from Sawyerpuram in Tinn- 

evelly District, South India. — Spolia zeylan 24(2), pp. 145-154. 

and MANLEY, F. P., 1942. The Manley Collection of 

Stone Age Tools. — Mem. archaeol. Surv. India, No. 68, 90 pp. 

ALLCHIN, B., 1952. A Study of some Palaeolithic artifacts from South 
India. — Current Sci. 21, pp. 268-271. 

ANDERSON, J., 1883. Catalogue and Handbook of the Archaeological 
Collections m the Indian Museum. — Part II, pp. 1-521. 

ANGLADE, A. and NEWTON, L. V., 1928. Dolmens of the Pulney Hills. 
Mem. archaeol Surv. India 36, pp. 1-13. 


BAIN, D. S. E., 1890. Mysore Megaliths. — J. anthrop. Soc. Bombay 2, 
pp. 229-31. 

BALL, V., 1875. On some stone implements of the Burmese type found in 
Pargana Dhalbhum district of Smghbhum, Chota-Nagpur division. — 
Proc. Asiatic Soc. Bengal 44, pp. 118-120. 

BECK, H. C, 1930. Notes on Sundry Asiatic Beads. — Man, London, 
October, No. 134. 

BOWLER-KELL.EY, A., 1937. Lower and Middle Palaeolithic facies in 
Europe and Africa. — 31 pp. — Philadephia (privately printed). 

BRANFILL, B. R., 1881(a). On the Savandurga rude-stone cemetery. 
Central Mysore. — Indian Antiq. 10, pp. 1-12. 

, 1881(b). Old slab-stone monuments in Madras and 

Maisur. — Indian Antiq, 10, 97-100. 

BROOKS, C. E. P., 1949. Climate through the Ages. — 395 pp. 2nd edit. — 

BROWN, C. J., 1917. Catalogue Raisonne of the Prehistoric Antiquities 
in the India Museum at Calcutta. — 155 pp. — Simla. 

BURKITT, M. C., 1920. Classification of Burins or Gravers. — Proc. prehist. 
Soc. East Anglia 3, pp. 306-11. 

, 1928. South Africa’s Past in Stone and Paint. — 183 pp. — 


CAMMIADE, L. A., 1924. Pigmy implements from the lower Godavari. — 
Man in India 4, pp. 83-105. 


CAMMIADE, L. A., 1930. Urn-burials in the Wynaad, South India. — Man, 
London, No. 135. 

and BURKITT, M.C., 1930. Fresh light on the Stone Ages 

in Southeast India. — Antiquity 4(15), pp. 327-339. 

CAMPBELL, T. D. and NOONE, H. V, V., 1943. South Australian Micro- 
lithic Stone Implements. — Rec. S. Austr. Mus. 7, pp. 281-307. 

CARNAC, R. H., 1883. On stone implements from the North Western 
Provinces of India. — J. Asiat. Soc. Beng. 52(1), pp. 221-230. 

CATON-THOMPSON, G., 1946. The Aterian Industry: Its place and 
significance in the Palaeolithic World. — J.R. anthrop. Inst. Huxley 
Mem. Lect. 76(2), pp. 87-130. 

CHAKRAVARTI, S. N., 1944. An outline of the Stone Age in India. — J. 
Asiat. Soc, Beng. (letters) 10, pp. 81-98. 

CHILDE, V. G., 1935. Changing methods and aims in Prehistory. — Presi- 
dent. Addr. Proc. prehist. Soc. 1, pp. 1-16. 

, 1942(a). What happened in History. — 256 pp. — New 


, 1942(b). The antiquity and function of antler axes and 

adzes. — Antiquity, London 16(63), pp. 258-264, 

, 1944. Archaeological ages as technological stages. — J.R. 

anthrop. Inst. Huxley Mem. Lect. 72, pp. 7-24. 

, 1946. Human Cultures as adaptations to environment — 

Geogr. J. 58(4/6), pp. 227-30. 

, 1947. Archaeology as a social science. — Rep. Univ. Lond. 

Inst. Archaeol. (1946) 3, pp. 49-60. 

, 1948. Megaliths — Ancient India No. 4 (July /Jan.) 

pp. 5-13. 

, 1950(a). The dawn of European Civilization — 5th edit. 

362 pp. — London. 

, 1950(b). Axe and Adze, Bow and Sling: Contrasts in 

Early Neolithic Europe. — Soc. Suisse prehist. 1949-50, pp. 156-162. 

CLARK, J. G. D., 1932 The Mesolithic Age in Britain. — 233 pp. — Cam- 

, 1934. The classification of a microlithic culture: the 

Tardenoisian of Horsham. — archaeol. J., London 90, pp. 52-77. 

, 1936. The Mesolithic Settlement of Northern Europe. — 

284 pp. — Cambridge. 

, 1944. Man and Nature in Prehistory with special ref- 
erence to Neolithic Settlement in Northern Europe. — IN — Occ. Pap. 
Univ. London. Inst. Archaeol. 6, pp. 20-8. 

, 1949. A preliminary report on excavations at Star Carr, 

Seamer, Scarborough, Yorkshire, 1949. — Proc. prehist. Soc. 15, 
pp. 50-2. 

, 1950. Preliminary report on excavations at Star Carr, 

Seamer, Scarborough, Yorkshire (Second Season, 1950). — Proc. pre- 
hist. Soc. 16, pp. 109-129. 


CLARK, J. G. D., 1952. Prehistoric Europe: The economic basis.— 349 pp. 
— London. 

and RANKINE, W.F., 1939. Excavations at Famham, 

Surrey (1937-38): The Horsham Culture and the question of Meso- 
lithic dwellings. — Proc. prehist. Soc. 5(1), pp. 61-118. 

CLIFFORD, E. M. and DANIEL, G. E., 1940. The Rodmarton and Aven- 
ing portholes. — Proc. prehist. Soc. 6(1), pp. 133-165. 

COCKBURN, J., 1879. Notes on stone implements from the Khasi Hills 
and the Banda and Vellor Districts. — J. Asiat. Soc. Beng. 48(2), 
pp. 133-143. 

CODRINGTON, K. de B., 1930. Indian Cairn and Urn-burials.— Man, 
London (Oct.) No. 139. 

COGHLAN, H. H., 1943. The evolution of the axe from Prehistoric to 
Roman times. — J.R. anthrop. Inst. 73, pp. 27-56. 

, 1951. Notes on the Prehistoric metallurgy of copper and 

bronze in the Old World. — Occ. Pap. Tech. Pitt Rivers Mus. 4, 
131 pp. 

COLE, R., 1873. Cromlechs in Mysore. — Indian Antiq. 2, pp. 86-88. 

DANIEL, G. E., 1941. The dual nature of the megalithic colonization of 
Prehistoric Europe. — Proc. prehist. Soc. 7, pp. 1-49. 

, 1943. The three ages: an essay on archaeological method. 

— 60 pp. — Cambridge. 

, 1950. The Prehistoric chamber tombs of England and 

Wales. — 256 pp. — Cambridge. 

DASGUPTA, H. C., 1913. On two shouldered stone-implements from 
Assam. — J. Asiat. Soc. Beng. (n.s.) 9. 

, 1931. Bibliography of Prehistoric Indian Antiquities. — 

J. Asiat. Soc. Beng. (n.s.) 27. 

DE TERRA, H., 1942. The megaliths of Burzahom, Kashmir, a new pre- 
historic civilization from India. — Proc. Amer. phil. Soc. 85, pp. 

and PATERSON, T. T., 1939, Studies on the Ice Age in 

India and associated human cultures. — Publ. Carneg. Instn. No. 493, 
354 pp. 

DIKSHITAR, V. R. R., 1940. Disposal of the dead in South India. — Proc. 
10th All-India Oriental Conf., pp. 530-33. 

DIXON, R. B., 1928. The building of cultures. — 312 pp. — New York. 

EICKSTEDT, Baron von, 1935. The position of Mysore in the racial 
history of India. — Mysore Tribes and Castes 1. 

EL WIN, V., 1951. The tribal art of Middle India. — 214 pp. 

FEBVRF, L„ 1950. A Geographical Introduction of History. — 3rd impr. 
388 pp. — London. 

FERGUSSON, J., 1872. Rude stone monuments in all countries; their age 
and uses. — London. 


FOOTE, R. B., 1873. On the geology of parts of the Madras and North 
Arcat districts lying north of the Palar river and included in sheet 
78 of the Indian Atlas. — Mem. Geol. Surv. India 10, 132 pp. 

, 1887. Notes on some recent Neolithic and Palaeolithic 

finds in South India. — J. Asiat. Soc. Beng. 56(2). 

, 1895. Geology of Bellary District, Madras Presidency. — 

Mem. Geol. Surv. India 25, 218 pp. 

— , 1901. Catalogue of the Prehistoric Antiquities of the 

Government Museum, Madras. — 131 pp. — Madras. 

, 1914. Indian Prehistoric and protohistoric Antiquities. — 

262 pp. — Madras. 

, 1916. Indian Prehistoric and Protohistoric Antiquities; 

notes on their age and distribution. — 245 pp. — Madras. 

GARROD, D. A. E., 1932. A new mesolithic industry: the Natufian of 
Palestine. — J.R. "anthrop. Inst. 62, pp. 257-269. 

, 1938. The Upper Palaeolithic in the light of recent dis- 
covery. — Proc. prehist. Soc. 4(1), pp. 1-26. 

, 1946. Environment, tools and Man. — 30 pp. — Cambridge. 

and BATE, D. M. A., 1937. The Stone Age of Mount 

Carmel. I. Excavations at the Wady El-Mughara. — 240 pp. — Oxford. 

GHOSH, A., 1948. Prehistoric exploration in India. — Indian Histor. Quart. 
24(1), pp. 1-18. 

GHOSH, M„ 1932. Rock paintings and other antiquities of prehistoric and 
later times. — Mem. archaeol. Surv. India 28, pp. 1-23. 

GHURYE, G., 1926. Funerary monuments in India. — Man in India 6, 
pp. 26-57, 100-138. 

GOBERT, E. G. and VAUFREY, R., 1932. Deux gisements extremes 
d’ Iberomaurusien —Anthrop., Paris. 42, pp. 449-490. 

GOODWIN, A. J. H., 1945. Method in Prehistory. — S. Afr. archaeol. 
Soc. Handb. 1, 191 pp. 

5 1946 The terminology of prehistory. — S. Afr. archaeol. 

Bull 1(4), pp 91-100. 

GORDON, D. H., 1938 The microlithic industries of India. — Man 38(19), 
pp. 21-24. 

, 1943 The cultures of Maski and Madhavpur. — J. Asiat. 

Soc. Beng. (Letters, 3rd ser.) 9, pp. 84-96. 

, 1945. Archaeological miscellany. — J. Indian anthrop. Inst. 

(n.s.) 1, pp. 28-30. 

, 1952. The early use of metals in India and Pakistan. — 

J.R. anthrop. Inst. 80(1/2), pp. 55-78. 

HAIMENDORF, C. von Furer, 1943. Megalithic ritual among Gadabas and 
Bondos of Orissa. — J. Asiat. Soc. Beng. (Letters, 3rd ser.) 9, pp. 

, 1945. The problem of Megalithic Cultures in Middle 

India. — Man in India 25, pp. 137-140. 

, 1948. Culture strata in the Deccan. — Man, London 48 

(89), pp. 87-90. 


HARGREAVES, H., 1929. Excavations in Baluchistan, 1925. Sampur 
Mound, Mastauig and Sohr Damb. — Mem. archaeol. Surv. India 35, 
pp. 1-89. 

HAYAV ADANA RAO, C., 1927. Mysore Gazetteer 1. 493 pp.— Bangalore. 

HUNT, E. H., 1916. Hyderabad Cairns. — J. Hyderabad archaeol. Soc., 
pp. 180-224. 

, 1924, Hyderabad Cairn Burials and their significance. — 

J.R. anthrop. Inst. 54, pp. 140-156. 

HUNTINGTON, E., 1924. Civilization and Climate.— New Haven. 

KELLEY, H., 1937. Acheulian flake tools.— Proc. prehist. Soc. 3(1), 
pp. 15-28. 

KENDRICK, T. D., 1925. The Axe Age. A study in British Prehistory. — 
177 pp. — London. 

KNOWLES, F. H. S., 1944. The manufacture of a flint arrowhead by a 
quartzite hammerstone. — Occ. Pap. Tech. Pitt Rivers Mus. 1, pp. 1-36. 

KRISHNA, M. H., 1929. Excavations at Chandravalli, Mysore State. — 
32 pp. 

, 1940. Mysore Archaeological Report. — pp. 63-74. 

, 1942(a). Excavations at Brahmagiri. — Mysore archaeol. 

Rep., pp. 100-109. 

} 1942(b). prehistoric Dakhan. — Proc. Indian Sci. CongT., 

pp. 223-252. 

KRISHNAN, M. S., 1949. Geology of India and Burma. — 544 pp. — Madras. 

KRISHNASWAMI, V. D., 1938(a). Environmental and cultural changes of 
prehistoric man near Madras. — J Madras gcogr. Ass. 13(1), pp. 58-90 

, 1938(b). Prehistoric Man round Madras. — Proc. Indian 

Acad. Sci. 4th Ann. Sess. Madras, pp. 86-111. 

, 1947. Stone Age India. — Ancient India No. 3, pp 11-49. 

LACAILLE, A. D., 1942. Scottish micro-burins. — Proc. Soc Antiquaries 
Scotland, 76, pp. 103-119. 

, 1947. The scraper in Prehistoric Culture. — Trans. Glas- 
gow archaeol. Soc. 11, pp. 38-93. 

LAF1TTE, N., 1931. A note on certain signs inscribed on pottery found in 
ancient funeral urns in South India. — Indian Antiq. 60, pp. 137-138. 

LAL, B. B., 1949. The chalcolithic phase in South Indian history. — J. 
Asiat. Soc. Beng. (Letters) 15(1), pp. 41-4. 

LEAKEY, L. S. B., 1936. Stone Age Africa. — 218 pp. — Oxford. 

, 1947. Palaeolithic nomenclature. — Man 47, pp. 19-20. 

, 1951. Preliminary excavations of a Mesolithic site at 

Abinger Common, Surrey. — Res. Pap. Surrey archaeol. Soc. No. 3, 
44 pp. 

LEEDS, E. T., 1920. The Dolmens and Megalthic Tombs of Spain and 
Portugal. — Archaeologia 70, pp. 201-232. 

LOGAN, A. C., 1906. Old chipped stones of India. — 85 pp. — Calcutta. 


LOWE, C. van Riet., 1945. The evolution of the Levallois Technique in 
South Africa. — Man, London 45, No. 37, pp. 49-59. 

MACKAY, E. J. H., 1937-38. Further excavations ait Mohenjo-Daro; being 
an official account of archaeological excavations at Mohenjo-Daro 
carried out by the Government of India between the years 1927-31. — 
2 vols., 718 pp. — New Delhi. 

MACKENZIE, J. S. F., 1873. On the rude stone archaeology of the Hassan 
District, Maisur. — Indian Antiq. 2, pp. 7-11. 

MALLOWAN, M. E. L. and ROSE, J. C., 1935. Prehistoric Assyria. 
The Excavations at Tall Arpachiyah 1933. — Iraq 2(1), 178 pp. 

MARSHALL, J., 1915. Excavations at Bhita. — Rep. archaeol. Surv. India, 
1911-12, pp. 29-94. 

, 1931. Mohenjo-Daro and the Indus Civilization. — 3 vols., 

716 pp. — London. 

MAZUMDAR, N. G., 1936. Museums. Section IV. Rep. archaeol. Surv. 
India 1930-1934, p. 259. 

MENON, K. G., 1937. Red painted pottery from Cochin State. — Man, 
London, No. 179. 

MOVIUS, H. L., 1944. Early man and Pleistocene stratigraphy in southern 
and eastern Asia. — Pap. Peabody Mus. 19(3), pp. 1-125. 

; 1949(a). Old World Palaeolithic Archaeology. — Bull. Geol. 

Soc. Amcr. 60, pp. 1443-1456. 

, 1949(b). The Lower Palaeolithic cultures of southern and 

eastern Asia. — Trans. Amer. phil. Soc. (1948) (n.s.) 38(4), pp. 330-420. 

NEWBOLD, Lieut., 1843. On the process prevailing among the Hindus, 
and formerly among the Egyptians, of quarrying and polishing granite. 
— J.R. Asiatic Soc. G.B.I., 7, pp. 113-128. 

NOONE, H. V. V., 1934. Burins and their classification. — J.R. anthrop. 
Inst. 44, pp. 81-92. 

5 1946. Australian microlithic industry. — Proc. prehist. Soc. 

12, pp. 173-174. 

NOONE, N. A. and H. V. V., 1940. The stone implements of Band- 
arawela. — Ceylon J. Sci. (G) 3(1). 

NORTH, F. J., 1938. Geology for Archaeologists. — Archaeol. J. (1937) 
94, pp. 73-115. 

OAKLEY, K. P., 1949. Man, the Tool-maker.— Publ. Brit. Mus. (Nat. 
Hist.) 98 pp. 

OLDHAM, R. D., 1893. Manual of the Geology of India. — 2nd edit. — 

PANCHAMUKHI, R. S., 1946. Dolmens and Cairns of Kamatak. — J. Univ. 
Bombay 14(IV), pp. 10-28. 

PATERSON, T. T., 1945. Core, culture and complex in the Old Stone 
Age. — Proc. prehist. Soc. 11, pp. 1-19. 


PEAKE, H. J. E., 1916. The Origin of the Dolmen.— Man, London, No. 68. 

, 1922. The Bronze Age and the Celtic World.— 201 pp.— 


PEET, T.E., 1912. Rough stone monuments and their builders. — 172 pp. — 

PERRY, W. J., 1915. The relation between the geographical distribution of 
megalithic monuments and ancient mines. — Mem. Manchr. lit. phil. 
Soc. 60(1), pp. 1-28. 

PLENDERLEITH, H. J., 1930. Black polished pottery from urn burials 
in the Wynaad. — Man, London (October) No. 138. 

PRADENNE, A. V., 1940. Prehistory. — 239 pp. — London. 

RAGHAVAN, M. D., 1935. Gourd-form in Pottery. — Man, London, No. 

RAMA RAO, B., 1935. Observations on some of the megalithic monuments 
in the vicinities of Halagur and Channapatra, Mysore State. — Indian 
Sci. Congr. 262 pp. 

, 1939. An introduction to the Geology of Mysore. — 65 pp. 

— Bangalore. 

, 1940. Archaean complex of Mysore. — Mysore Geol. De- 
part. Bull No. 17. 

RANKINE, W. F., 1951. A mesolithic survey of the West Surrey Green- 
sand. — Res. Pap. Surrey archaeol.. Soc. 2, pp. 1-50. 

REA, A., 1912. Prehistoric remains at Perambair. — Rep. archaeol. Surv. 
India for 1908-1909, pp. 92-99. 

, 1915. Catalogue of the Prehistoric Antiquities from 

Adicharallur and Perumbair — Madras. 

RICHARDS, F. J., 1924(a). Note on some Iron Age graves at Odugattur, 
North Arcot District, South India. — J R. anthrop. Inst. 54, pp. 157-165. 

, 1924(b). Anthropological Geography. — Man in India No. 4, 

pp. 14-40. 

, 1929. Cultural regions in India. — Geography 15. 

, 1933. Geographical factors in Indian Archaeology. — 

Indian Antiq. 62, pp. 235-243. 

, 1935. The cultural geography of Mysore. — Bangalore 

and CAMMIADE, L. A. and BURKITT, M.C., 1932. 

Climatic changes in southeast India during early Palaeolithic times. 
—Geol. Mag. 69, pp. 193-205. 

ROSS, E. J., 1946. A chalcolithic site in north-western Baluchistan. — J. 
Near Eastern Stud. 5, pp. 284-316. 

SAMPAT IYENGAR, P., 1924. A Palaeolithic settlement and factory in 
the Mysore State. — Indian Sci. Congr. 

, 1925. On some hand-grasp designs in Palaeoliths found in 

Mysore. — Indian Sci. Congr., p. 262. 

SANKALIA, H. D., 1946. Investigations into Prehistoric Archaeology of 
Gujarat. — 366 pp. — Baroda. 


SANKALIA, H. D., 1950. Ancient and Prehistoric Maharashtra. — J. Bombay 
Branch Asiat. Soc. 27(1), pp. 99-106, pis. 1-10. 

and KARVE, I., 1944. The 2nd Gujarat Prehistoric 

Expedition. — New Indian Antiquary 7(1), pp. 1-6. 

and , 1945. Preliminary Report on the 3rd 

Gujairat Prehistoric Expedition. — Deccan Coll. Res. Inst., Poona, 
16 pp. 

and , 1949. Early Primitive microlithic culture 

and people of Gujarat. — Amer. anthrop. 51(1), pp. 28-34. 

SKINNER, H. D., 1948. Chisel, wedge, axe and adze. — Antiquity 22, 


SLATER, H. K., 1906. Mysore Geological Department Records, 8, p.72. 

SMITH, G. Elliot, 1913. The origin of the Dolmen. — Man, London, 13 
(12), No. 105, pp. 193-5. 

SMITH, V. A., 1905. The copper age and prehistoric bronze implements 
of India. — Indian Antiq. 34, pp. 229-244. 

, 1906. Indian pigmy implements. — Indian Antiq. 35, 

pp. 185-95. 

SRINIVASAN, K. R., 1946. The Megalithic burials and urn-fields of South 
India in the light of Tamil literature and tradition. — Ancient India 
No. 2, pp. 9-16. 

SRIPADA RAO, K., 1930. On some stone implements from South India 
No. 1. — J. Mysore Univ. 4(2), pp. 1-8. 

, 1932. On some stone implements from South India No. 2. 

— Mysore univ. 6(1), pp. 1-12. 

SUBBA RAO, B., 1948. Stone Age Cultures of Bellary. — Deccan Coll. 
Dissertat. Ser. No. 7, pp. 1-62. 

TAYLOR, M., 1852. Megalithic tombs and other ancient remains in the 
Deccan. — Papers collected and published by the Archaeol. Depart. 
Hyderabad. 1941. 

THEOBALD, W., 1862. Celts found in Bundelkhand and chert implements 
from the Andamans. — Proc. Asiat. Soc. Beng. 31, pp. 323-27. 

TODD, K. R. U., 1932. Prehistoric Man around Bombay. — Proc. prehist. 
Soc. East. Angl. 7(1), pp. 35-42. 

, 1939. Palaeolithic industries of Bombay. — J.R. anthrop. 

Inst. 69, pp. 257-272. 

, 1948. A microlithic industry in Eastern Mysore. — Man, 

London, 48, No. 27, pp. 28-30. 

VAUFREY, R., 1933. Notes sur le Capsien. — Anthropologie, Paris 43, 
pp. 457-483. 

VAVILOV, N., 1926. Studies on the origin of cultivated plants. — Bull. 
Applied Bot., 16(2), 153 pp. 

VOGT, E., 1949. The Birch as a source of raw material during the Stone 
Age. — Proc. prehist. Soc. 15, pp. 50-51. 


WADIA, D, N., 1926. Geology of India for students, — 400 pp. — London. 

WAEGHTER, J. and SETON-WILLIAMS, V., 1938. Excavations at Wadi 
Dhobai 1937-38 and the Dhobaian industry.— J. Palestine Orient. 
Soc. 18 , 23 pp. 

WALHOUSE, M. J., 1874. Holed dolmens.— Indian Antiq. 3, p. 277. 

, 1875. Miniature and prehistoric pottery. — Indian Antiq. 

4, pp. 12-13. 

WALSH, E. H. C., 1904. A note on stone implements found in the Dar- 
jeeling District. — J. Asiat. Soc. Beng. 73(3), pp. 20-24. 

WARMINGTON, E. H., 1928 Commerce between the Roman Empire and 
India. 147 pp. — Cambridge. 

WATKINS, A. E., 1933. The Origins of Cultivated Plants. — Antiquity 7, 
pp. 73-80. 

WAYLAND, E. J. and BURKITT, M. C., 1932 The Magosian Culture 
of Uganda. — J.R. anthrop. Inst. 62, pp. 369-390. 

WETHERELL, E. W., 1901. Mysore Geological Department Records. — 
Vol. 3 (1900-1901). 

WHEELER, R. E. M., 1946. Archaeology in India Today. — Adhare- 
handra Mookerjee Lectures for 1944. Calcutta Univ., 29 pp 

, 1948. Brahmagiri and Chandravalh, 1947: Megalithic and 

other cultures in the Chitaldrug District, Mysore State. Ancient 

India No. 4 (July, 1947 - Jan., 1948), pp. 181-310. 

WORMAN, E. C, 1949. The “ neolithic ” problem in the Prehistory of 
India.— Proc. Wash Acad. Set. 39(6), pp. 181-201. 

WRIGHT, W.B., 1939. Tools and the Man —236 pp.— London. 

YAZDANI, G., 1917. The megalithic remains of the Deccan: a new aspect 
of them. — J. Hyderabad archaeol. Soc., pp. 56-79. 

ZEUNER, F. E., 1945. Review of the chronology of the Palaeolithic Period. 
— IN Univ. Lond.Inst. Archaeol. Occ. Pap. No. 6, pp. 14-19. 

s 1948(a) The Exhibition of Stone Age and Pleistocene 

Geology from the Cape to Britain. — Occ Pap. Univ. Inst. Archaeol. 
No. 9, 63 pp. 

, 1948(b). Climate and Early Man in Kenya. — Man, Lon- 
don 48 , No. 14, pp. 13-16. 

, 1950. Stone Age and Pleistocene chronology in Gujarat. — 

Deccan Coll. Monogr. Ser. No. 6, 46 pp. 

, 1952(a). Dating the Past. — 3rd edit. 495 pp. — London. 

, 1952(b). The Microlithic industry of Langhnaj, Gujarat. 

— Man, 50, No. 18, pp. 129-131. 

, and ALLCHIN, B., 1956. The Microlithic Sites of the 

Tinnevelly District. — Ancient India (in the press). 



PLATE I. Map showing die distribution of rocks in Mysore State and 
adjacent parts of the Deccan. Mysore boundary marked by a 
thick line (after Rama Rao). 

PLATE II. Distribution of Trap in northern Mysore and Bellary. 

PLATE III. Distribution of Palaeolithic and MicTolithic sites in Mysore. 
PLATE IV.A. Distribution of Brahmagiri Microlithic industry IA and IB. 

Key to sites: 

1 . 








































B. Distribution of Russet-coated Ware from Megalithic burials. 

PLATE V. Distribution of Polished Stone Axes and Megalithic sites in 

PLATE VI. Palaeolithic implements from Kibbanahalli, Mysore (£ nat. 

Figs. 1, 2, 3 and 7. Cleavers. 

Figs. 4, 5 and 6. Hand-axes. 

Figs. 8 and 9. Beaked tools. 

Fig. 10. Pointed flake. 

Fig. 11. Point. 

PLATE VII. Palaeolithic implements from Kibbanahalli, Mysore (i nat. 

Figs. 1 and 2. Pointed flakes or borers. 

Fig. 3. Disc. 

Fig. 4. Chopping-tool. 

Fig. 5. Chopper. 


Fig. 6. Pointed flake. 

Figs. 7 and 8. Scrapers. 

Fig. 9. Fabricator. 

Figs. 10 and 11. Cores. 

PLATE VIII. Palaeolithic implements from Mysore State compared with 

those from other Indian localities (i nat. size). 

Fig. 1. Hand-axe (No. 211) from Nidaghatta. 

Figs. 2 and 6. Hand-axe (No. 207) and chopper (No. 208) from 

Figs. 3, 4, 5, 8 and 10. Hand-axes (Nos. 217, 219, 220, 213 
and 216, respectively) from Lingadahalli. 

Fig. 7. Ovate (No. 202) from Karadigudda. 

Fig. 9. Ovate (No. 215) from Lingadahalli. 

Fig. 11. Hand-axe (No. AV III, 5) from Vadamambakkam, near 
Arkonam, Madras Province. 

Fig. 12. Hand-axe (No. AT III, 51) from Attirampakkam. 
Madras Province. 

PLATE IX. Palaeolithic artifacts from Mysore State compared with those 

from other Indian sites (1 nat. size). 

Figs. 1 and 2. Disc -like chopping-tools from Lingadahalli (No. 
222 and Attirampakkam, Madras Province (No. AT 88), re- 

Figs. 3 and 4. Hand-axes from Talya (No. 204) and Attirampak- 
kam, Madras Province (No. AT III, 38), respectively. 

Figs. 5 and 6. Hand-axes from Kadur (No. 226) and Attirampak- 
kam, Madras Province (No. AT III, 32), respectively. 

Figs. 7 and 8. Irregular discs from Talya (No. 206) and Attiram- 
pakkam, Madras Province (No. AT II, 67), respectively. 

Figs. 9 and 10. Hand-axes from Lingadahalli (No. 214) and 
Perumucch, four miles south of Arkonam. Madras Province 
(No. AV V, 2), respectively. 

Figs. 11 and 12. Bifacial pebble-tools from Nyamati (No. 225) 
and Chintaladevi, Nellore District (No. AV XIV, 10), 

PLATE X. Palaeolithic artifacts from Mysore State compared with those 

from other Indian sites nat. size). 

Figs. 1, 2 and 3. Ovates from Talya, Mysore (No. 205), Uliyam- 
bakkam, near Arkonam, Madras Province (No. AV VII, 3) 
and Attirampakkam, Madras Province (No. AT V, 65), respec- 

Figs. 4, 5 and 6. Discs from Nidaghatta, Mysore State (No. 209), 
Kilkuppam, near Arkonam (No. AV II, 17) and Ichchiputtam, 
near Arkonam, Madras Province (No. AV I, 4), respectively. 

Figs. 7 and 8. Hand-axes from Lingadahalli, Mysore (No. 212) 
and Attirampakkam, Madras Province (No. AT III, 54), re- 


Figs. 9 and 10. Choppers from Lingadahalli, Mysore (No. 218) 
and Vadamambakkam, Madras Province (No. AV III, 3), 

Figs. 11 and 12. Ovates from Lingadahalli, Mysore (No. 221) and 
Attirampakkam, Madras (No. AT II, 172), respectively. 

PLATE XI. Palaeolithic artifacts from Mysore State compared with those 

from other Indian sites (} nat. size). 

Figs. 1 and 2. Discs from Nidaghatta, Mysore (No. 210) and 
Attirampakkam, Madras Province (No. AT I, 143), respectively. 

Figs. 3 and 4. Hand-axes from Jodikatte, Mysore (No. 203) and 
Kilkuppam, near Arkonam, Madras Province (No. AV II, 16), 

Figs. 5 and 6. Ovates on flakes from Lingadahalli, Mysore (No. 
223) and Attirampakkam, Madras Province (No. AT III, 191), 

Figs. 7 and 8. Chopping-tools from Nyamati, Mysore (No. 224) 
and Khyad, Dharva District (No. 32), respectively. 

PLATE XII. A. Fabrication of microliths. The notch method. (After 
Rankine, 1951). 

B. Rejuvenation of microlithic cores: — 

1. Lateral trimming. 

2. Apical trimming. 

3. Basal trimming. 

(After Rankine, 1951). 

C. Reconstruction of the use of microliths as arrowheads. 
(After Leakey, 1950). 

PLATE XIII. Microliths from Jalahalli, Mysore State, collected by Todd 

(Coll. British Museum). 

Figs. 1-8. Petits tranchets and tranchet-like lunares. 

Figs. 9-40. Points blunted on the left. 

Figs. 41-46. Points blunted on the right. 

Figs. 47-51. Drills. 

Figs. 52-58. Triangles. 

Figs. 59-62. Burins. 

Figs. 63-66. Blades. 

The following numbers preceded by T refer to Fig. 2 in Todd 
(1948) and the original specimens that could be located and 
drawn for the present publication are indicated. 

T1 = Fig. 4 T28 = Fig. 53 T43 = Fig. 27 

T2 = „ 3 T29 = „ 51 T44 = „ 49 

T3 = „ 5 T32 = „ 25 T49 = „ 66 

T5 = „ 1 T33 = „ 26 T50 = „ 64 

T6 = „ 6 T34 = „ 55 T51 = „ 65 

T7 = „ 2 T35 = „ 13 T52 = „ 63 

T9 = „ 7 T38 = „ 54 T53 = „ 59 

T27 = „ 52 T42 = „ 47 


PLATE XIV. Microliths from Jalahalli, Mysore State, collected by Todd 
(Coll. British Museum). 

A. Scrapers. 

B. Lunates. 

The following numbers preceded by T refer to Fig. 2 in Todd 
(1948) and the orig inal specimens that could be located and 
drawn for the present publication are indicated. 

T18 = Fig. B7 
T25 = „ B4 
T55 = „ A10 
T56 = „ A2 

T57 = „ A8 
T58 = „ A13 

PLATE XV. Microliths from Bangalore, Mysore State. 

Figs. 1 and 3-15. Points. 

Fig. 2. Irregular triangles. 

Figs. 16-19. Lunates. 

All from Bangalore-Sarjapur Road. 

Figs. 20-30. From the Hindustan Aircraft Factory Site. 

PLATE XVI. Scrapers from the site near the Bangalore-Sarjapur Road, 

PLATE XVII. Microliths from Brahmagiri, Mystore State. Surface finds 
made by the writer. 

A. Pre-I industry. Figs. 1-7. Scrapers and used flakes. 

B. IA-B industries. 

Figs. 1-5 and 10-12. Parallel-sided blades, 2 and 4 with use- 

Fig. 6. Blunted-back blade. 

Fig. 7. Point. 

Fig. 8. Core. 

Fig. 9. Bladelet. 

PLATE XVIII. Microliths from Kibbanahalh, Mysore State. 

PLATE XIX. Polished stone implements from Brahmagiri, Mysore State 
(i nat. size). 

Fig. 1. Small adze. 

Fig. 2. Chisel. 

Fig. 3. Axe of la type. Surface find made by the writer. Clas- 
sified as adze/axe in this publication. 

PLATE XX. Polished stone axes from Brahmagiri. Surface finds made 
by the writer, (i nat. size). 

Fig. 1. Illustrating Stages C and D (retouching and pecking) in 
the manufacture of polished stone axes. 

Fig. 2. Axe resembling Wheeler’s type Alla, but with very 
rounded cutting edge. 

Fig. 3. Axe resembling Wheeler’s type Bla. 


Geological sketch map. 


Plate III 




/ ; 
c ,_ *JOU*vV<».UWUKM f 

, V > ^WwRGA \ 


SHWoC k 

OHJIPKTTf^ ckiv«M><J*GA. ^ ' 

&UDNKPuR*^rtk*&^xi - 

x y« — < i „■*/ > 

Ctf'' > jr H'i • cuO\8ANdV 

) <t / O'^AN&lA ; 

j / \ •PERESANtSRA , 


f M1KHACALUR / , ■ NANDI -, 

) m l .tWuWKVAOt KARM.AKA'ITE 6»\ X W A *>APU»A A ~ 

NALe^. \ • 9DMGAMT O.,^ A^ * •n.ANA.O»D»««.AU^ 

•Konomii • •bAmuiomia-.J 

AGOMlftm aOHMKpp^na UAUL» *Ol>R. Pkv-VA* 

t H *SSAH * H A V U # MlOOAW AM »aU-» 

■ ■ M WHKfiUHOA-»* 

h *Lt^» \ *puv 

• X ACONlgtD . 

t m^ssah 

\ WSOQft* 

*&fcn*viu.| y 

M&UD^Jk # J 

^niNCHROtHj ^ 

•HAlCaCU^ ) 

-> UHHAP PkM A UALL1 »OU*. . 



SMAWADUtOA ^ s ~ ^ r r 


/ ^ 

l_* CAS tf^ 

— v», 

>0 4rO So 

4 Pousmo SfOHE AoilC 



Plate IV 

IA 6 IB P so too /s o Mfles 

Plate V 


Plate XVI 

Polished stone implements from Brahmagiri (} nat size)